A material management system based on machine learning
Through the material management system based on machine learning, real-time monitoring and matching of materials and engineering projects, the problem of uncontrollable material quality inspection time in the existing technology has been solved, and the on-time delivery of material transportation and the progress of engineering projects have been achieved.
Patent Information
- Application Number
- CN202410969066.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-19
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-07-19
AI Technical Summary
The existing material management and control system lacks real-time monitoring of the quality inspection information of engineering projects, resulting in uncontrollable quality inspection time and prone to delay construction periods.
The material management system based on machine learning is adopted to obtain materials and project information through the information acquisition module, and the data matching module performs intelligent matching. The progress reminder module monitors the quality inspection progress in real time, generates a late warning, and ensures that the materials are delivered on time.
It improves the real-time nature of the material transportation process, reduces the delay in construction period, and promotes the implementation progress of the project.
Smart Images

Figure CN118504953B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of material control technology, and in particular to a material management system based on machine learning. Background Art
[0002] In the existing technology, the material management and control system mainly covers the entire life cycle of material procurement, production (self-made parts processing), recycling and reuse of companies, teams, etc. From production planning, to planned procurement (storage and sales on behalf of others), warehousing and outbound delivery, production consumption (including recycling), evaluation and assessment, etc., the whole process and multi-faceted management and control are realized, and the actual operation process of the company, material planning, procurement, contract, storage, collection, recycling and reuse, financial and budget control, etc. is integrated and centralized. Management and supervision.
[0003] However, the existing management and control system lacks information on the corresponding engineering projects for the tested materials. The staff of the material quality inspection party are not clear about the usage time requirements of the corresponding engineering projects for the materials to be tested. The project liaison person is also unclear about the quality inspection queue status of the project materials and whether the time required for quality inspection will exceed the expected start time. The construction period is easily delayed due to the uncontrollable quality inspection time. Summary of the invention
[0004] The present application provides a material management system based on machine learning, which enables the staff at the material quality inspection end to control the quality inspection efficiency in real time according to the construction time in the material project information, and the staff at the project docking end to monitor the estimated quality inspection end time in the material project information in real time, which greatly promotes the implementation progress of various engineering projects and reduces delays in construction period.
[0005] The present application embodiment provides a material management system based on machine learning, including:
[0006] The information acquisition module is used to obtain material information and project information; material information includes the estimated end time of quality inspection and material supplier; project information includes construction time and contact information of the project docking end;
[0007] The data matching module is used to input material information or project information into the machine learning model for matching to obtain material project information; and send the material project information to the material quality inspection end and the project docking end respectively;
[0008] The progress reminder module is used to determine whether the difference between the estimated quality inspection end time and the construction time in the material project information is less than the preset logistics time; if so, a material late arrival warning is generated and sent to the material quality inspection end.
[0009] Furthermore, the system also includes:
[0010] The detection and analysis module is used to use the big data platform to analyze historical inspection results, determine the inferior suppliers and the corresponding material project information, and send inferior quality reminder information to the project docking end of each material project information.
[0011] Furthermore, the system also includes:
[0012] The cloud storage module is used to store the material information, material inspection results and project information of the information acquisition module, as well as the material project information of the data matching module and the inferior suppliers of the detection and analysis module.
[0013] Furthermore, the data matching module is also used to train the machine learning model using the data in the cloud storage module according to a preset optimization cycle, so that it matches the corresponding project information according to the material information, or matches the corresponding material information according to the project information.
[0014] Furthermore, the system also includes:
[0015] The data query module is used to receive query instructions and determine whether the query instructions hit the preset sensitive word database; if not, search in the cloud storage module according to the query instructions, obtain the query results and display them.
[0016] Furthermore, the system also includes:
[0017] The login verification module is used to query the received user name and password based on the account database; after querying the user name and password, a blacklist or whitelist is used to perform security verification on the user's IP address; after the security verification is passed, the existing session window is deleted and a new session window is created with a random length.
[0018] Furthermore, the system also includes:
[0019] The operation and maintenance module is used to delete the session window after the session window exists for more than a preset closing time.
[0020] Furthermore, the system also includes:
[0021] The data sensitivity determination module is used to sample and extract data from the host to obtain message data; an asynchronous analysis algorithm is used to identify sensitive data and sensitivity levels of message data to form a sensitive database;
[0022] The interface classification module is used to obtain the historical traffic data of the data interface, determine the sensitivity level of the historical traffic data based on the sensitive database, and use it as the sensitivity level of the data interface;
[0023] The interface alarm module is used to obtain real-time traffic data of the data interface and determine the current sensitivity level of the real-time traffic data based on the sensitive database; and to issue an alarm when the current sensitivity level exceeds the sensitive level of the data interface.
[0024] Furthermore, the system also includes:
[0025] The baseline building module is used to obtain the historical operation data of the same user name, and to model and learn the historical operation data to obtain the normal behavior baseline of the user name;
[0026] The user alarm module is used to obtain the current operation behavior of the user name, extract the abnormal behavior that exceeds the normal behavior baseline and calculate the user risk value, and issue an alarm when the user risk value exceeds the preset behavior threshold.
[0027] Furthermore, the normal behavior baseline includes a normal access frequency range, a normal access time range, a normal access data range, and a normal login time interval.
[0028] In summary, compared with the prior art, the technical solution provided in the embodiment of the present application has at least the following beneficial effects:
[0029] The embodiment of the present application provides a material management system based on machine learning. First, the information acquisition module is used to acquire material information, material inspection results and project information from multiple terminals. Then, the data matching module uses a machine learning model to intelligently match the material information and the corresponding project information. After the material project information is obtained, it is sent to the material quality inspection terminal and the project docking terminal respectively, thereby improving the real-time monitoring of the material transportation process, so that the staff at the material quality inspection terminal can control the quality inspection efficiency in real time according to the construction time in the material project information, and the staff at the project docking terminal can monitor the estimated quality inspection end time in the material project information in real time. The progress reminder module realizes real-time process monitoring and reminders to ensure the timely delivery of materials, which greatly promotes the implementation progress of various engineering projects and reduces delays in construction period. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1 A structural diagram of a material management system based on machine learning is provided as an exemplary embodiment of the present application.
[0031] Figure 2 A structural diagram of a material management system based on machine learning is provided as another exemplary embodiment of the present application.
[0032] Figure 3 A structural diagram of a material management system based on machine learning is provided as yet another exemplary embodiment of the present application. DETAILED DESCRIPTION
[0033] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.
[0034] Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of this application.
[0035] Terminology explanation:
[0036] (1) Supplies: Supplies include equipment, materials, furniture and others. Equipment, materials and others support the requisition operation, others refer to unclassified types of materials, and furniture supports the allocation operation.
[0037] (2) Sensitive words: usually refer to words that are set to require special attention or be automatically filtered in specific software, applications or services. For example, words that involve political issues or may cause controversy, words that violate laws and regulations or contain illegal content, words that involve security vulnerabilities or are used for fraud, words that involve business secrets or sensitive information of competitors, etc.
[0038] (3) Blacklist: A list of prohibited or disallowed entries. Entries on the blacklist, such as IP addresses, email addresses, user accounts, etc., will be automatically denied access or certain operations by the system. Blacklists are often used to prevent malicious users, spammers, or known bad actors from accessing the system.
[0039] The advantages of blacklists are that they can quickly identify and block known bad behaviors; they are easy to update and manage. The disadvantage is that they may not be able to respond to new threats because blacklists rely on known bad entries. In network security, blacklists can be used to block access requests from known malicious IP addresses; email systems may use blacklists to filter out emails from known spammers; in user rights management, blacklists can list users who are prohibited from performing certain operations; in content management systems, blacklists can be used to filter out inappropriate words or content.
[0040] (4) Whitelist: A list of allowed or trusted entries. Only entries on the whitelist can obtain access rights or perform operations. Other entries not listed will be denied. Whitelists are often used to create a secure environment that only allows access from known secure or trusted sources. The advantage of a whitelist is that it provides high security because only verified entries are allowed, reducing the possibility of false positives. The disadvantage is that it is too strict and limits flexibility; it needs to be constantly updated to include new trusted entries.
[0041] In network security, whitelists can ensure that only requests from specific IP addresses are allowed. Email systems might use whitelists to ensure that messages from specific senders are always accepted. In user rights management, whitelists can list users who are specifically allowed to perform those actions. In content management systems, whitelists can ensure that certain content is always allowed to be published.
[0042] (5) API call: API (Application Programming Interface) call is a way for software systems to interact with each other. API is a set of predefined functions, protocols and tools for building software applications. API call refers to the process of a program sending a request through API and receiving a response. API call usually starts with a request, which includes the operation to be performed and any necessary parameters or data. The endpoint of API is a specific URL on the server that handles the request. Each API call points to a specific endpoint. API calls can use different HTTP methods, such as GET (get data), POST (submit data), PUT (update data), DELETE (delete data), etc. The response of API call usually contains the execution result or returned data. The response format may be JSON, XML or other data format.
[0043] (6) Web interface: refers to the user interface of an application accessed through the Internet. This interface allows users to interact with the application through a web browser. Users can access the web interface through any device with a browser, regardless of the operating system or device type. The web interface is not dependent on a specific operating system and can work on platforms such as Windows, macOS, Linux, iOS, Android, etc. The web interface can dynamically load and display content without the user having to download additional software or applications. Modern web interfaces typically include forms, buttons, drop-down menus, and other interactive elements that allow users to interact with web pages. The development of web interfaces usually follows open standards such as HTML, CSS, and JavaScript, and can protect user data and privacy through secure protocols such as HTTPS.
[0044] (7) Session window: A session window usually refers to an interface element for users to interact with a business system. It allows users to communicate and operate with the system in real time through a graphical user interface (GUI), allowing users to input information, select options, or execute commands, and see instant feedback from the system. The session window can handle multiple tasks or queries at the same time, and display information provided by the business system, such as data entry forms, search results, reports, charts, etc. Users can enter data or select options through interface elements such as text boxes, drop-down menus, and buttons. The session window may contain navigation elements that enable users to jump between different functional modules or pages. The session window can display system status, such as a loading indicator, a waiting message, or a processing progress bar.
[0045] See also Figure 1 , the embodiment of the present application provides a material management system based on machine learning, including:
[0046] The information acquisition module is used to obtain material information and project information; material information includes the expected completion time of quality inspection and the material supplier; project information includes the construction time and the contact information of the project docking end.
[0047] Among them, the data source of the estimated quality inspection end time in the material information is the material quality inspection end, the data source of the material supplier, material number, material size and name, material quantity and other information in the material information is the material allocation management platform, and the data source of the project information is the project docking end; the information acquisition module obtains the above information in real time.
[0048] The data matching module is used to input material information or project information into the machine learning model for matching to obtain material project information; and send the material project information to the material quality inspection end and the project docking end respectively.
[0049] Specifically, when the information acquisition module obtains material information or project information, the data matching module inputs it into the trained machine learning model, matches the corresponding project information or material information, and merges them to form material project information.
[0050] This application uses machine learning technology to intelligently integrate and summarize the material information and project information collected from the data platform, achieving efficient merging of materials and projects while ensuring the validity of the information.
[0051] First, according to the business needs of the material management system, extract or construct features from the original data that are helpful for the machine learning model to understand, which may include the classification of materials, key indicators of the project, etc.; then integrate the original data to form a unified data set, select a suitable machine learning model, such as a classification model, regression model, clustering model, etc., use the integrated data set to train the machine learning model, and evaluate the performance of the machine learning model through methods such as cross-validation, analyze the importance of each feature in the model, and understand which features are most critical to the model's prediction; finally, adjust the model parameters or select a different model based on the performance of the model to improve the accuracy and efficiency of the prediction.
[0052] The progress reminder module is used to determine whether the difference between the estimated quality inspection end time and the construction time in the material project information is less than the preset logistics time; if so, a material late arrival warning is generated and sent to the material quality inspection end.
[0053] Among them, the material late arrival warning includes the material project information, so that the material quality inspection end can find the corresponding materials according to the material information in the material project information, speed up its quality inspection, and update the estimated quality inspection end time in the material project information; after receiving the updated material project information, this application cancels the warning after judgment by the progress reminder module.
[0054] The preset logistics time is the time required for the material quality inspection end to transport the materials to the project construction site. The progress reminder module uses this as the alarm standard, which is conducive to the materials needed for the project to be delivered to the construction site more quickly, thereby achieving the lean standard of material management and control.
[0055] The above embodiment provides a material management system based on machine learning. First, the information acquisition module acquires material information, material inspection results and project information from multiple terminals, and then uses the machine learning model to intelligently match the material information and the corresponding project information through the data matching module. After the material project information is obtained, it is sent to the material quality inspection terminal and the project docking terminal respectively, thereby improving the real-time monitoring of the material transportation process, so that the staff at the material quality inspection terminal can control the quality inspection efficiency in real time according to the construction time in the material project information, and the staff at the project docking terminal can monitor the estimated quality inspection end time in the material project information in real time. The progress reminder module realizes real-time process monitoring and reminders to ensure the timely delivery of materials, which greatly promotes the implementation progress of various engineering projects and reduces delays in construction period.
[0056] See also Figure 2 In some embodiments, the system further comprises:
[0057] The detection and analysis module is used to use the big data platform to analyze historical inspection results, determine the inferior suppliers and the corresponding material project information, and send inferior quality reminder information to the project docking end of each material project information.
[0058] The information acquisition module is also used to obtain the material inspection results from the material quality inspection end. The material inspection results correspond to the material information one by one, but the upload time of the material inspection results is later than the time of obtaining the material information.
[0059] Specifically, in the existing material management and control system, the project liaison person can only determine whether the materials can be put into use after obtaining the material inspection results of the project. If the quality inspection fails, a new order must be placed. However, in fact, the materials of different projects may come from the same supplier, and different project liaisons cannot communicate with each other, resulting in the materials of inferior suppliers being repeatedly inspected by different projects, which not only wastes material inspection resources, but also causes the project to be delayed due to re-ordering materials.
[0060] Therefore, this application uses big data to analyze past material inspection results through the information acquisition module. The big data platform can be the QuickBI platform, which focuses on screening out suppliers whose unqualified materials exceed a certain proportion as inferior suppliers, and promptly notifies the project docking person who has ordered the inferior supplier based on the matched material project information, so that the project docking end can make a timely decision on whether to return the goods. Furthermore, this application can also mark suppliers whose qualified rate of historical inspection results exceeds a certain proportion as high-quality, so that the project docking end can make choices when ordering materials and ensure a better level of material quality control.
[0061] In some embodiments, the system further comprises:
[0062] The cloud storage module is used to store the material information, material inspection results and project information of the information acquisition module, as well as the material project information of the data matching module and the inferior suppliers of the detection and analysis module.
[0063] The cloud storage used in this application is a service that stores data on a remote server, and users can access this data via the Internet. Cloud storage services are usually provided by third-party providers, and storage capacity can be expanded on demand. Compared with traditional local storage, cloud storage is usually more economical because users only pay for the storage space used. And cloud storage services usually provide automatic backup and data recovery functions to prevent data loss or damage. Cloud storage service providers usually also take a variety of security measures to protect user data, including encryption and access control, to further ensure the security of data.
[0064] In the specific implementation process, the cloud storage resources used in this application are:
[0065]
[0066] In some embodiments, the data matching module is also used to train a machine learning model using data in the cloud storage module according to a preset optimization cycle, so that it matches the corresponding project information according to the material information, or matches the corresponding material information according to the project information.
[0067] Specifically, the present application implements a feedback mechanism of the machine learning model through a cloud storage module, that is, during the application process, the matching capability of the machine learning model is optimized according to a preset optimization cycle, thereby continuously improving the matching accuracy and speed.
[0068] For example, the cloud storage module already stores material information and corresponding project information, as well as matched material project information. The material information in the cloud storage module is input into the machine learning model, and the material project information corresponding to the material information is used as the supervision to adjust the model parameters of the machine learning model; further, when the machine learning model matches incorrectly, the staff at the project docking end can feedback the error message after receiving the material project information and provide the correct material information. This application requires the data matching module to input the material information feedback from the project docking end and the project information in the erroneous material project information into the machine learning model after receiving the error message, so that it can be adjusted according to the correct matching relationship.
[0069] See also Figure 3 In some embodiments, the system further comprises:
[0070] The data query module is used to receive query instructions and determine whether the query instructions hit the preset sensitive word database; if not, search in the cloud storage module according to the query instructions, obtain the query results and display them.
[0071] Among them, the query command input by the user can be information such as material number, material name, project contact person name, etc. The query command can be initiated through API call, Web interface or other methods, and the display and initiation methods of the query results are the same.
[0072] Specifically, this application can prevent users from using query string parameters that contain sensitive words or affect the server security logic by presetting a sensitive word database. Furthermore, the cloud storage module can encrypt data before storing it, for example, using the transparent data encryption (TDE) function or encrypting data at the application level. Specific sensitive fields such as passwords and personal information can be encrypted using strong encryption algorithms (such as AES).
[0073] The data query module is also used to implement strict access control policies to ensure that only authorized users can perform queries and access data. In order to avoid SQL injection attacks by malicious users, parameterized queries are used to process user input. When returning query results, only necessary data is returned to avoid unnecessary exposure of sensitive data.
[0074] In some embodiments, the system further comprises:
[0075] The login verification module is used to query the received user name and password based on the account database; after querying the user name and password, a blacklist or whitelist is used to perform security verification on the user's IP address; after the security verification is passed, the existing session window is deleted and a new session window is created with a random length.
[0076] Among them, the login verification module performs security verification on the user's IP address, which can prevent illegal IP access and address spoofing.
[0077] The random length of the session window can prevent attackers from guessing it. The login verification module also stores the session data generated after the user successfully logs in on the server, and sets access verification conditions for the session data on the server based on the user's login information to ensure that the session data cannot be illegally accessed, and logs out the session data stored on the server when the user logs out.
[0078] During the specific implementation process, since the material management system connects multiple platforms and coordinates a large amount of data, its security and privacy are also key issues that each user must consider before accessing it. Existing material management and control systems pay more attention to information shielding between multiple user terminals, and pay less attention to malicious intrusions by unfamiliar users. However, in actual applications, as the number of users of the material management and control system increases, malicious attacks will also increase accordingly. Therefore, how to protect the multi-party data in the system from being tampered with or stolen is also an issue that the material management system of this application needs to focus on and solve.
[0079] The above embodiment not only performs security verification on the IP address of the logged-in user through the login verification module, but also changes the session data after the user logs in to a random length, which can better protect the security of system data and user session data.
[0080] In some embodiments, the system further comprises:
[0081] The operation and maintenance module is used to delete the session window after the session window exists for more than a preset closing time.
[0082] Among them, the preset shutdown time can be set to 20 minutes.
[0083] Furthermore, the operation and maintenance module of the present application is also used to prohibit users from using configuration files in the system's Web space to prevent possible server configuration vulnerabilities from causing configuration files to be downloaded, and at the same time, the user's modification, deletion, and change of access rights to the configuration files are recorded in the system log. The system log of the present application uses the log operation package Log4j program provided by Apache to output messages such as the start and stop of the system, system execution information, and error logs to the log log of the application server. In addition, the system of the present application also provides an access log to record each user's login behavior and access process, including information such as user ID, operation type, operation time, session flag, and access menu.
[0084] The above embodiment sets the session window to be closed at a scheduled time, which can reduce the risk of session hijacking attacks, reduce the occupation of system resources by long-suspended sessions, and improve system performance; in addition, if the user forgets to close the session, the scheduled closing can prevent data from being leaked in the background and reduce the risk of unauthorized access.
[0085] In some embodiments, the system further comprises:
[0086] The data sensitivity determination module is used to sample and extract data from the host to obtain message data; an asynchronous analysis algorithm is used to identify sensitive data and sensitivity levels of message data to form a sensitive database.
[0087] Among them, the asynchronous analysis algorithm includes regular expressions, natural language processing, knowledge base, column names, text similarity, data fingerprints, file attributes and other technical means; in addition to sampling the data of the host, it can also sample and extract data from various databases, big data platforms and other storage media and perform comparative analysis, and determine the sensitive data and sensitivity level of the received message data according to the relevant requirements of the superior regulatory authorities or the enterprise for the classification and grading management of sensitive system data, thereby constructing a sensitive database including various sensitive data and corresponding sensitivity levels, providing data support for subsequent data security threats and risk analysis.
[0088] Specifically, the regular algorithm will create a corresponding regular expression based on the received sensitive data type definition. For example, a regular expression for matching email addresses may be [a-zA-Z0-9_.+-]+@[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+; then, the regular expression is used to scan the message data to find and identify the sensitive data therein.
[0089] The calculation method of text similarity can include a variety of techniques, such as Levenshtein distance: measuring the minimum number of editing operations required to convert two texts from one to another, including inserting, deleting and replacing characters; Longest Common Subsequence (LCS): characterizing the similarity between texts by calculating the length of the longest common subtext of two texts; N-Gram: by setting a sliding window of size N, forming multiple text fragments of length N, calculating the ratio of the number of common N-Grams in two texts to the total number of N-Grams to characterize text similarity; Jaccard similarity: based on the idea of sets, calculating the ratio of the number of elements in the intersection and union of two sets, suitable for texts that are not sensitive to the order of characters / words; Cosine similarity: inspired by the law of cosines, measuring the similarity between texts by calculating the cosine value of the text vector, taking into account the frequency of the text; Simhash: sensitive hashing algorithm, mapping documents to a binary vector, calculating the similarity between texts through Hamming distance, suitable for judging whether the content of documents is the same.
[0090] Data Fingerprinting can generate a unique identifier for each message data. By comparing data fingerprints, the sensitivity level of message data can be divided according to the different attributes of the fingerprints.
[0091] The interface classification module is used to obtain the historical traffic data of the data interface, determine the sensitivity level of the historical traffic data based on the sensitive database, and use it as the sensitivity level of the data interface.
[0092] The interface alarm module is used to obtain real-time traffic data of the data interface and determine the current sensitivity level of the real-time traffic data based on the sensitive database; and to issue an alarm when the current sensitivity level exceeds the sensitive level of the data interface.
[0093] This application records the access track and content of the data interface based on traffic collection, including the IP address of the interface access party, the access duration of the interface access party, the business type of the interface access party and the system. The access frequency of the interface during working hours, the access frequency of the interface during non-working hours, the interface call time interval, the amount of data transmitted by the interface, the classification of data transmitted by the interface, the field of data transmitted by the interface, etc., and defines the sensitivity level of the interface according to the sensitivity level of these traffic data.
[0094] After completing the definition of the sensitivity level of the interface, the real-time traffic data is monitored according to the sensitivity level, and an alarm is issued when the current sensitivity level of the real-time traffic data exceeds the sensitivity level of the data interface.
[0095] The above-mentioned embodiment realizes real-time monitoring and alarm of system data interface, which can reduce the risk of data leakage, maintain business continuity and stability, and prevent data security incidents from affecting the business; in addition, when a user attempts to access or transmit sensitive data that exceeds the sensitivity level of the data interface, the alarm can prevent possible misoperation.
[0096] In some embodiments, the system further comprises:
[0097] The baseline building module is used to obtain the historical operation data of the same user name, and model the historical operation data to obtain the normal behavior baseline of the user name. The normal behavior baseline may include the normal access frequency range, the normal access time range, the normal access data range and the normal login time interval.
[0098] Specifically, before modeling and learning, the source IP, destination IP, destination port, account number, and access location in the historical operation data are combined into five-tuple information, and the five-tuple information is used as a data set to train and learn the model to obtain a normal behavior baseline.
[0099] The user alarm module is used to obtain the current operation behavior of the user name, extract the abnormal behavior that exceeds the normal behavior baseline and calculate the user risk value, and issue an alarm when the user risk value exceeds the preset behavior threshold.
[0100] Among them, abnormal behaviors may include abnormal frequency of access to sensitive data, abnormal frequency of fuzzy query operations on sensitive data, abnormal frequency of designated business operations, abnormal access time, abnormal access during non-working hours, access to sensitive data without logging in for a long time, multiple people using the same account, simultaneous logins or changes in the IP address of the login client, abnormal operation methods and habits, abnormal changes in access to data carriers, abnormal access to data scope, and abnormal changes in access to sensitive businesses.
[0101] Specifically, the steps for calculating the user risk value include: counting all normal behaviors to obtain a rule set G = {g1, g2, ..., gn}, where gn is the nth rule; counting abnormal behaviors to obtain a rule occurrence set C = {c1, c2, ..., cm}; comparing the rule occurrence set C with the rule set G to obtain the rule occurrence ratio, and multiplying the rule occurrence ratio by 100 to obtain the user risk value.
[0102] The above-mentioned embodiments reduce the need for manual monitoring by real-time monitoring and warning of user operation behaviors, and can promptly identify abnormal users or malicious users, thereby ensuring the security of the system and reducing operational risks.
[0103] In some embodiments, the present application also includes a risk monitoring module, which is used to determine the assessment level according to the interface type or user risk value of the alarm data interface after the system alarm, monitor the alarm data interface or user according to the assessment level, trace the source and collect evidence based on the monitored traffic data or current operation behavior, and form corresponding risk warning information.
[0104] Specifically, the evaluation levels include the first-level evaluation level, the second-level evaluation level and the third-level evaluation level. The evaluation level is determined based on the importance of the interface, the severity of the fault type or the size of the user risk value. The first-level evaluation level, the second-level evaluation level and the third-level evaluation level represent the degree of danger of abnormal behavior from low to high.
[0105] For example, when the alarm interface is an Ethernet interface, the assessment level is the third-layer assessment level; when the alarm interface is a fiber optic interface, a serial interface, or a parallel interface, the assessment level is the second-layer assessment level; when the alarm interface is a satellite interface, Zigbee, or Z-Wave interface, the assessment level is the first-layer assessment level.
[0106] When the user risk value is between 0 and 30, the assessment level is the first-level assessment level; when the user risk value is between 31 and 70, the assessment level is the second-level assessment level; when the user risk value is between 71 and 100, the assessment level is the third-level assessment level.
[0107] Different assessment levels are only reflected in the risk warning information, and no different treatment is required for monitoring and traceability.
[0108] In some embodiments, the present application also includes a situational awareness module for monitoring periodic changes in data resources, characterizing different data boundaries according to different business function attributes, and monitoring data exposure and unauthorized operation behaviors.
[0109] This application can play a role in business combing and resource processing through the situation awareness module, and can analyze and respond to risks to assist in decision-making.
[0110] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. The technicians in the field can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the material management system and modules described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.
[0111] In the embodiments disclosed herein, it should be understood that the disclosed products (including but not limited to systems, devices, etc.) can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, the indirect coupling or communication connection of the modules can be electrical, mechanical or other forms. The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, the functional modules in the various embodiments of the present invention can be integrated into a processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.
[0112] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0113] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.
Claims
1. A material management system based on machine learning, characterized in that: include: An information acquisition module is used to acquire material information and project information; the material information includes the estimated end time of quality inspection and the material supplier; the project information includes the construction time and the contact information of the project docking end; A data matching module is used to input the material information and the project information into a machine learning model for matching to obtain material project information; and send the material project information to a material quality inspection end and a project docking end respectively; The progress reminder module is used to determine whether the difference between the estimated quality inspection end time in the material project information and the construction time is less than the preset logistics time; if so, a material late arrival warning is generated and sent to the material quality inspection end.
2. The material management system based on machine learning according to claim 1 is characterized in that: Also includes: The detection and analysis module is used to use the big data platform to analyze historical inspection results, determine the inferior suppliers and the corresponding information of each material project, and send inferior quality reminder information to the project docking end of each material project information.
3. The material management system based on machine learning according to claim 2 is characterized in that: Also includes: A cloud storage module is used to store the material information and the project information of the information acquisition module, the material project information of the data matching module and the inferior supplier of the detection and analysis module.
4. The material management system based on machine learning according to claim 3 is characterized in that: The data matching module is also used to train the machine learning model using the data in the cloud storage module according to a preset optimization cycle, so that it matches the corresponding project information according to the material information, or matches the corresponding material information according to the project information.
5. The material management system based on machine learning according to claim 3 is characterized in that: Also includes: A data query module, used to receive a query instruction and determine whether the query instruction hits a preset sensitive word database; If there is no hit, a search is performed in the cloud storage module according to the query instruction to obtain and display the query result.
6. The material management system based on machine learning according to claim 1, characterized in that: Also includes: A login verification module is used to query the received username and password based on the account database; After the user name and the password are queried, a blacklist or a whitelist is used to perform security verification on the user's IP address; after the security verification is passed, the existing session window is deleted, and a new session window is created with a random length.
7. The material management system based on machine learning according to claim 6 is characterized in that: Also includes: The operation and maintenance module is used to delete the session window after the session window exists for more than a preset closing time.
8. The material management system based on machine learning according to claim 1, characterized in that: Also includes: The data sensitization module is used to sample and extract data from the host to obtain message data; Using an asynchronous analysis algorithm to identify sensitive data and sensitivity levels of the message data to form a sensitive database; An interface classification module, used for obtaining historical traffic data of a data interface, determining a sensitivity level of the historical traffic data based on the sensitive database, and using the sensitivity level of the data interface as the sensitivity level; The interface alarm module is used to obtain the real-time traffic data of the data interface, determine the current sensitivity level of the real-time traffic data based on the sensitive database, and issue an alarm when the current sensitivity level exceeds the sensitive level of the data interface.
9. The material management system based on machine learning according to claim 6, characterized in that: Also includes: A baseline building module is used to obtain historical operation data of the same user name, and to perform modeling and learning on the historical operation data to obtain a normal behavior baseline of the user name; The user alarm module is used to obtain the current operation behavior of the user name, extract the abnormal behavior that exceeds the normal behavior baseline and calculate the user risk value, and issue an alarm when the user risk value exceeds a preset behavior threshold.
10. The material management system based on machine learning according to claim 9, characterized in that: The normal behavior baseline includes a normal access frequency range, a normal access time range, a normal access data range and a normal login time interval.
Citation Information
Patent Citations
Material agreement inventory demand prediction method based on hybrid model
CN111401808A
Building construction optimization system based on big data and cloud computing
CN116862199A
Water conservancy construction project material management system
CN118037182A
System and method for managing transportation assets
US20030135304A1
System and method for allocating the supply of critical material components and manufacturing capacity
WO2002060235A2
