CN116993504A

CN116993504A - Steel transaction information service platform and transaction data processing method

Info

Publication number: CN116993504A
Application number: CN202311237321.9A
Authority: CN
Inventors: 张易; 邝锋; 叶祖焕; 熊大为
Original assignee: Hunan Valin E Commerce Co ltd
Current assignee: Hunan Valin E Commerce Co ltd
Priority date: 2023-09-25
Filing date: 2023-09-25
Publication date: 2023-11-03
Anticipated expiration: 2043-09-25
Also published as: CN116993504B

Abstract

The invention relates to the technical field of communication, in particular to a steel transaction information service platform and a transaction data processing method, wherein the method comprises the following steps: acquiring real-time transaction data of a first steel product of a user; real-time data acquisition is carried out on the steel by using a sensor; performing feature extraction on the multi-mode steel real-time data to generate steel transaction information feature data; carrying out data integration on the characteristic data of the steel transaction information; performing interactive view processing on the steel transaction information characteristic database; performing expansion convolution and multi-scale sampling on the steel real-time interactive view; carrying out data binding on the steel characteristic convolution model and the first transaction data of the user; performing data dimension reduction on a user real-time steel transaction model; the invention realizes the efficient processing and real-time management of the steel transaction data.

Description

Steel transaction information service platform and transaction data processing method

Technical Field

The invention relates to the technical field of communication, in particular to a steel transaction information service platform and a transaction data processing method.

Background

The traditional steel transaction data processing method is to store the data in a series of handwriting documents, manual records and other modes, the data updating is not enough in time, the data processing efficiency is low, the transaction data is usually stored in a form of a spreadsheet or a text file, and many challenges are faced due to the fact that the transaction data rely on manual operation and simple tools. The challenges include speed and accuracy of data input and efficiency and expandability of data processing, so an intelligent steel transaction information service platform is needed, all data are recorded and integrated on the data processing service platform in real time in a factory site through the steel transaction information service platform and a transaction data processing method, data processing and data analysis flow of steel transaction information can be greatly simplified, data processing efficiency and quality are improved through an association rule learning algorithm, perfect steel transaction data processing and data service can be provided for two parties of steel transaction through combining technologies such as artificial intelligence and cloud computing, real-time data updating of steel transaction information is achieved, market demands are better met through intelligent algorithm analysis and real-time model construction, and more accurate, rapid and intelligent data processing and analysis service is provided.

Disclosure of Invention

The invention provides a steel transaction information service platform and a transaction data processing method for solving at least one technical problem.

In order to achieve the above object, the present invention provides a steel transaction data processing method, comprising the steps of:

step S1: acquiring real-time transaction data of a first steel product of a user by utilizing an information acquisition module; acquiring real-time data of the steel by using a sensor to acquire real-time data of the multi-mode steel; the multi-mode steel real-time data comprise steel real-time video, steel price data, steel material data, steel origin data, steel specification grade data, steel mass density data and steel real-time transaction condition data;

step S2: carrying out feature extraction on the multi-mode steel real-time data by utilizing a feature engineering algorithm to generate steel transaction information feature data; carrying out data integration on the steel transaction information characteristic data by utilizing a knowledge graph method to generate a steel transaction information characteristic database;

step S3: performing interactive view processing on the steel transaction information feature database by using an association rule learning algorithm to generate a steel feature real-time interactive view;

Step S4: performing expansion convolution and multi-scale sampling on the steel characteristic real-time interactable view by utilizing a convolution neural network to construct a steel characteristic convolution model;

step S5: carrying out data binding on the steel characteristic convolution model and the first steel real-time transaction data of the user by utilizing a bidirectional binding algorithm to generate a real-time steel transaction model of the user;

step S6: performing data dimension reduction on the user real-time steel transaction model by using a principal component analysis method to generate second steel real-time transaction data of the user;

step S7: symmetrically encrypting the second steel real-time transaction data of the user by utilizing a symmetrical encryption algorithm to generate encrypted data of the steel real-time transaction of the user; and uploading the encrypted data of the user real-time steel transaction to the steel transaction information service platform by using a linear programming method, so as to realize steel transaction data management.

The invention can acquire the real-time transaction data of the first steel of the user and the real-time data of the multi-mode steel in real time through the information acquisition module and the sensor. The data comprise information such as videos, prices, materials, origins, specification grades, quality densities and the like, accurate and real-time steel transaction data can be provided, users are helped to know the current market conditions, the multi-mode data enriches the expression form of the transaction data, more dimensional information is provided, feature extraction and data integration are carried out on the multi-mode steel real-time data through a feature engineering algorithm and a knowledge graph method, representative features can be extracted, data dimensions are reduced, important information is reserved, the data integration helps integrate data from different sources together, more comprehensive steel transaction information is obtained, the association rule learning algorithm is utilized to carry out interactive view processing on a steel transaction information feature database, the steel feature real-time interactive view is generated, feature modes and structure information in the view can be extracted, the steel feature convolutional model is generated through the learning of a convolutional neural network, the data real-time steel transaction model is carried out through a main component analysis method, second steel real-time transaction data of users is generated, complexity of the data is reduced through the dimension reduction, processing and analysis efficiency is improved, the most important transaction features are extracted, more visual and important transaction information are provided for users, the transaction information is smoothly transmitted to a user, the user transaction information is encrypted by the aid of the symmetric steel transaction algorithm, the data is managed by the data, the user data is encrypted, the security and the user transaction information is encrypted, the data is encrypted, the security is ensured, the data is managed, the security and the user transaction information is encrypted, the data is encrypted, and the data is safe and is ensured.

Preferably, step S1 comprises the steps of:

step S11: acquiring user first steel real-time transaction data by using an information acquisition module, wherein the user first steel real-time transaction data comprises user purchase data, user transaction amount data, user identity information data, user transaction time stamp and user transaction state data;

step S12: acquiring real-time images of steel by using a video sensor to generate steel real-time videos;

step S13: detecting the material quality of the steel by using an ultrasonic sensor to obtain steel material quality data, steel specification grade data and steel mass density data;

step S14: and acquiring steel price data, steel origin data and steel real-time transaction condition data by using the steel transaction information service platform.

According to the invention, the first steel real-time transaction data of the user is obtained through the information acquisition module, accurate user transaction related data is provided, the purchasing behavior and transaction condition of the user can be tracked, user identity information data can be used for identity verification and safety management, transaction time stamp and transaction state data provide time sequence information and transaction state tracking, a video sensor is utilized to acquire real-time images of the steel, steel real-time videos are generated, visual information is provided, the user can observe the appearance and quality of the steel through the videos, the videos can be used for quality inspection, quality verification and recording, an ultrasonic sensor is utilized to detect the quality of the steel, steel material data, steel specification grade data and steel quality density data are obtained, detailed material information of the steel is provided, the data comprises data such as material type, strength grade, density and the like, the data can be used for quality control, material selection and performance evaluation, a steel transaction information service platform is utilized to obtain steel price data, steel origin data and steel real-time transaction condition data, market price information of the steel is provided, the user is helped to know market price trend, origin data provide origin information of the product, and the user can consider geographical factors and supply chain management conditions, and transaction opportunity and transaction risk can be used for analyzing real-time market risk and transaction conditions. By the steps, accurate user transaction data, visual steel information (video), accurate material data, market price information, origin data and real-time transaction condition data can be provided. Such information may assist the user in making informed decisions, understanding market dynamics, selecting appropriate steel materials and specifications, and enhancing the reliability and transparency of steel transactions.

Preferably, step S2 comprises the steps of:

step S21: performing feature detection on the multi-mode steel real-time data by using a feature point detection algorithm to generate a steel transaction information feature code;

step S22: carrying out feature extraction on the steel transaction information feature codes by using a feature engineering method to generate steel transaction information feature data;

step S23: node element mining is carried out on the steel transaction information characteristic data by utilizing a knowledge graph method, and steel transaction information concept node data is generated;

step S24: node relation combination is carried out on the steel transaction information concept node data, and a steel transaction information node relation data set is generated;

step S25: and carrying out data integration on the steel transaction information node relation data set to generate a steel transaction information characteristic database.

The invention carries out feature detection on the multi-mode steel real-time data through a feature point detection algorithm to generate steel transaction information feature codes, the feature point detection algorithm can identify key feature points in the steel real-time data, such as edges, angular points and the like, the generated feature codes can encode the feature point information, the processing and characterization of unstructured data of steel transaction information are provided, feature engineering is utilized to carry out feature extraction on the steel transaction information feature codes to generate steel transaction information feature data, the feature engineering can extract features with representativeness and degree of distinction from original codes so as to facilitate subsequent data processing and analysis, the generated feature data has higher information value and interpretability, can be used for data mining, machine learning and model establishment, node element mining is carried out on the steel transaction information feature data by utilizing a knowledge graph method to generate steel transaction information concept node data, the knowledge graph method can mine concept nodes from the feature data, namely data units with semantic association and association, the generated concept node data can provide higher data abstraction and semantic association, can be favorable for carrying out more integrated relation analysis on steel transaction information and steel transaction information with integrated relation between nodes and data in a more comprehensive relation set of node analysis and data can be formed, the relation between the transaction information and the node analysis can be integrated, and the relation node relation can be integrated, the relation can be formed to the relation node analysis can be more comprehensively formed, the method comprises the steps of generating a steel transaction information feature database, integrating data of different sources and different formats by data integration to form an integrated feature database, wherein the generated feature database can provide comprehensive steel transaction information features including structured data, unstructured data, semantic association and the like, provide richer resources for further data analysis, model establishment and decision support, provide deeper and comprehensive understanding of steel transaction information, and provide more powerful support for the fields of data analysis, data mining, decision support and the like.

Preferably, step S23 comprises the steps of:

step S241: performing concept attribute labeling on the steel transaction information concept node data to generate steel transaction information concept attribute data;

step S242: performing conceptual relation mining on the steel transaction information conceptual attribute data based on a lexical rule method to generate steel transaction information conceptual relation data;

step S243: and carrying out node relation combination on the steel transaction information concept relation data to generate a steel transaction information node relation data set.

According to the method, the concept attribute marking is carried out on the steel transaction information concept node data to generate the steel transaction information concept attribute data, the concept attribute marking can be used for clearly marking the meaning and attribute of the concept nodes, so that the meaning and attribute of the concept nodes are clearer and more specific on a semantic level, the generated concept attribute data provides description and definition of different attributes of the steel transaction information, understanding and analysis of the characteristics and meaning of the concept nodes are facilitated, the concept relation mining is carried out on the steel transaction information concept node data based on a lexical rule method to generate the steel transaction information concept relation data, the lexical rule method can be used for finding out the association relation between different concept nodes in the steel transaction information through means such as text analysis and association mining, the generated concept relation data can provide interaction and dependency relation between the concept nodes, the disclosure of steel transaction flow and relevant influence factors are facilitated, the node relation association is carried out on the steel transaction information concept relation data to generate a steel transaction information node relation data set, the node relation association can integrate and integrate the relation between different concept nodes, the generated node relation data set can comprise rich association information, the meaning and relation between different concept nodes can be better understood and analyzed, the meaning and the analysis is better and more fully provided, the analysis is provided for the analysis of the concept is better understood and more complete, and the field is better analyzed.

Preferably, step S3 comprises the steps of:

step S31: frequent item set mining is carried out on the steel transaction information feature database by utilizing an association rule learning algorithm, and a steel transaction information feature vector is generated;

step S32: performing data visualization processing on the steel transaction information feature vector to generate a steel transaction information feature visualization view;

step S33: and carrying out interactive processing on the visual view of the steel transaction information characteristic by using a JavaScript library to generate a real-time interactive view of the steel characteristic.

According to the invention, frequent item set mining is carried out on the steel transaction information feature database through the association rule learning algorithm, so as to generate the steel transaction information feature vector, the association rule learning algorithm can analyze the frequent item set in large-scale data, help find out the features which frequently occur simultaneously in the steel transaction information, such as specific product combination or transaction conditions, the generated steel transaction information feature vector can convert the transaction information into numerical representation, subsequent analysis and modeling are convenient, higher usability is provided for pattern recognition, prediction or classification tasks and the like, data visualization processing is carried out on the steel transaction information feature vector, so as to generate a steel transaction information feature visualization view, the data visualization processing can display the steel transaction information feature vector in a chart, graph or other visual forms, help people to understand and analyze data more intuitively, the generated feature visualization view can display the relation, distribution situation and change trend among features, the understanding capability of global view and local detail of the steel transaction information is provided, the steel transaction information feature visualization view is subjected to interactive processing by the JavaScript library, the steel transaction information feature visualization view is generated, the steel feature visualization view is subjected to real-time interaction view is provided, the JavaScript interaction processing is provided, the user interaction view is provided with a visual interaction view, and the interaction operation view is provided with a user-defined interaction view, and the interaction view is more visible, and the user interaction view is provided with the user-defined and has the interaction view, and has the interaction function, and is better-defined, and is provided by the interaction view, and has the interaction function.

Preferably, step S4 comprises the steps of:

step S41: performing convolution preprocessing on the steel characteristic real-time interactable view by using a convolution neural network to generate a steel transaction information characteristic sample set;

step S42: performing convolution data cutting on the steel transaction information feature sample set by using a super-pixel algorithm to generate a steel transaction information convolution feature sequence;

step S43: performing edge feature reinforcement processing on the steel transaction information convolution feature sequence by using an expansion convolution algorithm to generate a steel transaction information convolution feature network;

step S44: carrying out spatial pyramid pooling multi-layer sampling on the steel transaction information convolution characteristic network by utilizing a multi-scale sampling algorithm to generate a steel transaction information convolution characteristic diagram;

step S45: performing data mining algorithm modeling on the steel transaction information convolution feature map by utilizing a combined classifier weighting comprehensive calculation formula based on a combined classifier algorithm to generate a steel transaction information convolution feature model;

according to the invention, the convolution neural network is used for carrying out convolution pretreatment on the steel characteristic real-time interactable view to generate the steel transaction information characteristic sample set, the convolution neural network can be used for extracting the characteristics in the image data, the convolution pretreatment is carried out on the steel characteristic real-time interactable view to capture the key characteristics in the view, the generated steel transaction information characteristic sample set contains the image characteristics subjected to the convolution pretreatment, input data is provided for the subsequent steps, the steel transaction information characteristic sample set is subjected to convolution data cutting by utilizing the super-pixel algorithm to generate the steel transaction information convolution characteristic sequence, the super-pixel algorithm can be used for dividing the image into continuous super-pixel areas with similar characteristics, the super-pixel cutting is carried out on the steel transaction information characteristic sample set to convert the sample set into a series of small blocks with similar characteristics, the generated steel transaction information convolution characteristic sequence can help to retain local characteristics and context information, more detailed and rich characteristic representation is provided, the edge characteristic reinforcement processing is carried out on the steel transaction information convolution characteristic sequence by utilizing the expansion convolution algorithm to generate the steel transaction information convolution characteristic network, the expansion convolution algorithm can be used for expanding the wild feeling of the convolution core, and the edge feeling capability of the edge information is enhanced. By applying the expansion convolution algorithm, the steel transaction information convolution feature network can better capture edge features, the characterization capability of a model is improved, the generated steel transaction information convolution feature network has richer and higher-level feature representation, the steel transaction information convolution feature network is better differentiated into different types of steel transaction information, the multi-scale sampling algorithm is utilized to carry out spatial pyramid pooling multi-layer sampling on the steel transaction information convolution feature network, a steel transaction information convolution feature map is generated, the multi-scale sampling algorithm can pay attention to features of different scales at the same time, the spatial pyramid pooling multi-layer sampling is carried out on the steel transaction information convolution feature network, feature information of different scales can be captured, the generated steel transaction information convolution feature map can provide rich local and global perception, the subsequent data mining and analysis are facilitated, the steel transaction information convolution feature map is modeled by utilizing a combined classifier weighting comprehensive calculation formula to generate a steel transaction information convolution feature model, the steel transaction information convolution feature model can be fused based on the combined classifier algorithm, the combined classifier weighting comprehensive calculation is carried out on the steel transaction information convolution feature map, the predicted feature map can be obtained, and the new transaction information classification and the data mining and classification task classification can be facilitated.

Preferably, the combined classifier weighted comprehensive calculation formula in step S45 is specifically:

；

wherein ,for combining classifier weight coefficients, +.>Is->Personal classifier, < >>For the number of basis classifiers, +.>Is the firstWeights of the individual basis classifier in the combined classifier,/->Is->Predictive value of weight by the individual basis classifier, < +.>For inputting sample values of the initial basis classifier, < +.>Sum of predicted results for the result values for the basis classifier,/->Is->Personal basePrediction result of classifier on result value, +.>Is->Accuracy of the prediction of the radix classifier, +.>For the number of classification results, +.>Is->The personal classifier pair->Weights of the individual basis classifier, +.>Is->Classifying results of the individual basis classifier on the sample, < >>For sample->In->The value of the individual basis classifier.

The invention is realized byRepresenting the sum of the base classifier weight coefficients multiplied by the predicted value of each base classifier pair weight. By weighted summing the predicted values of the individual base classifiers, the degree of contribution of each base classifier to the final combined classifier can be captured. This is beneficial for taking into account the predictive power of the different base classifiers in combination. By passing throughRepresenting the product of the base classifier's predicted outcome and the number of classified outcomes for the outcome value multiplied by and square root divided. It may consider a balanced relationship between the number of classification results and the predicted results of the base classifier on the result values. This is beneficial in adjusting the final weight according to the number of classification results and the predictive accuracy of each classification result. By- >The Sigmoid function maps the predicted values of the base classifier weights, converting the predicted values into probability values ranging from 0 to 1. This is beneficial to normalize the predictors so that they can represent the degree of weight impact of the individual basis classifiers and can be interpreted and understood by means of probability values. By->Representing the classification result of the base classifier on the samples multiplied by the weights between the base classifiers. This may allow for further adjustment of the final weights taking into account interactions and dependencies between different base classifiers. The formula calculates and adjusts the weight coefficient of the combined classifier by comprehensively considering the factors such as weight prediction, result prediction, accuracy, quantity and sample classification result of the base classifier. This is beneficial to more fully evaluating the contribution of the base classifier, balancing the different factors, and further improving the performance and accuracy of the combined classifier.

Preferably, step S5 comprises the steps of:

step S51: presetting a binding frame by utilizing a bidirectional binding algorithm to a steel characteristic convolution model to generate a steel transaction bidirectional binding frame;

step S52: carrying out data frame binding on the steel characteristic convolution model and the first steel real-time transaction data of the user by utilizing a steel transaction bidirectional binding frame to generate a steel transaction data frame of the user;

Step S53: and updating the real-time data of the user steel transaction data frame by using the data monitor to generate a user real-time steel transaction model.

The invention uses the bidirectional binding algorithm to preset the binding frame for the steel characteristic convolution model to generate the steel transaction bidirectional binding frame, the bidirectional binding algorithm can establish a real-time data binding relation between the steel characteristic convolution model and transaction data to realize bidirectional transmission and synchronous updating of the data, the generated steel transaction bidirectional binding frame can provide clear data flow and communication mechanism, simplify the complexity of data processing and model updating, uses the steel transaction bidirectional binding frame to bind the steel characteristic convolution model with the first transaction data of the user to generate the steel transaction data frame of the user, the steel transaction bidirectional binding frame can seamlessly connect the steel characteristic convolution model with the first transaction data of the user, the method comprises the steps that a frame binding relation of data is established, the generated user steel transaction data frame can integrate and display steel transaction data of a user, a more convenient and consistent data operation and access mode are provided, the data monitor is utilized to conduct real-time data updating on the user steel transaction data frame, a user real-time steel transaction model is generated, the data monitor can monitor changes of the user steel transaction data frame and update related data information in real time, the generated user real-time steel transaction model can be updated in real time along with updating of the user transaction data, current and accurate steel transaction information is provided, and a more efficient, real-time and reliable steel transaction model is provided, so that the user can track, analyze and manage steel transaction data better.

Preferably, step S6 comprises the steps of:

step S61: performing data dimension reduction on the user real-time steel transaction model by using a principal component analysis method to generate user real-time steel transaction standardized data;

step S62: performing matrix division on the standardized data of the user real-time steel transaction by using a self-adaptive division method to generate a data matrix of the user real-time steel transaction;

step S63: non-negative matrix factorization is carried out on the real-time steel transaction data matrix of the user by using a matrix factorization method, so that a main component matrix of the real-time steel transaction of the user is generated;

step S64: and calculating eigenvalues of the principal component matrix of the user real-time steel transaction by using a covariance matrix algorithm, and generating second steel real-time transaction data of the user.

The invention utilizes the principal component analysis method to carry out data dimension reduction on the user real-time steel transaction model to generate the user real-time steel transaction standardized data, the principal component analysis method can extract the most representative characteristic from the original data, the high-dimensional data is reduced to a data space with lower dimension, the generated real-time steel transaction standardized data has lower dimension, the complexity and redundant information of the data are reduced, the data processing efficiency and the accuracy of the model are improved, the user real-time steel transaction standardized data are divided by utilizing the adaptive division method to generate the user real-time steel transaction data matrix, the adaptive division method can divide the standardized data according to a certain rule to form a proper data matrix structure, the generated real-time steel transaction data matrix can better organize and manage the data, the convenience is provided for the subsequent analysis and calculation, the non-negative matrix decomposition is carried out on the user real-time steel transaction data matrix by utilizing the matrix decomposition method to generate the user real-time steel transaction main component matrix, the non-negative sub-matrix is decomposed, the non-negative characteristic and sparse characteristic of the original data are reserved, the generated main component analysis matrix is more compact, the real-time steel transaction main component data is more represented by utilizing the principal component analysis matrix and the redundant component analysis data, the principal component analysis method is more convenient, the principal component analysis and the real-time steel transaction data has better characteristic analysis value and the principal component analysis matrix is calculated, the principal component analysis and the principal component analysis data has better characteristic analysis value and the real-time analysis value is calculated by utilizing the principal component analysis algorithm of the principal component analysis data, the generated real-time transaction data of the second steel contains characteristic information with higher importance, so that the distribution, trend and variability of the data can be known, and a foundation is provided for decision and prediction.

Preferably, step S7 comprises the steps of:

step S71: performing data ciphertext conversion on the second steel real-time transaction data of the user by using a symmetric encryption algorithm to generate a steel real-time transaction symmetric data ciphertext;

step S72: symmetric encryption is carried out on symmetric data ciphertext of the steel real-time transaction by utilizing a symmetric encryption calculation formula of the steel real-time transaction information, so as to generate symmetric encryption data of the steel real-time transaction;

step S73: performing network scheduling slicing on the steel real-time transaction symmetric encryption data by using a linear programming method to generate a plurality of steel real-time transaction symmetric encryption data;

step S74: and uploading the encrypted data of the steel transaction of the user in real time to a steel transaction information service platform to realize steel transaction data management.

The invention uses symmetric encryption algorithm to conduct data cryptogram conversion on the second steel real-time transaction data of the user to generate steel real-time transaction symmetric data cryptogram, the symmetric encryption algorithm can encrypt and decrypt the data by using the same secret key to ensure confidentiality and security of the data, the generated steel real-time transaction symmetric data cryptogram converts the original data into an encrypted form to prevent sensitive information from being accessed by unauthorized individuals or entities, the steel real-time transaction symmetric data cryptogram is symmetrically encrypted by using the steel real-time transaction information symmetric encryption calculation formula to generate steel real-time transaction symmetric encryption data, the steel real-time transaction information symmetric encryption calculation formula can conduct further encryption operation on the data cryptogram according to a specific algorithm and secret key, the generated steel real-time transaction symmetric encryption data enhances the security of the data, the unauthorized parties are prevented from analyzing and acquiring data content, the steel real-time transaction symmetric encryption data is subjected to network scheduling slicing by using a linear programming method, the generated multiple steel real-time transaction symmetric encryption data can be sliced and scheduled by an optimized method, the data are distributed to different network resources, the generated multiple steel transaction symmetric encryption data can improve the efficiency of data transmission and processing, and reduce the load of the data transaction data can be transmitted to a steel transaction service platform, the data can be conveniently encrypted by a user service platform, the data can be shared by the transaction service, the data can be conveniently and the transaction service can be managed and data can be transmitted to the data platform, the data transaction service can be encrypted and has good stability, and data can be managed and data can be encrypted and managed and can be shared by a service and encrypted, the processing and analyzing functions support the user to perform more efficient and accurate data management and transaction operation, are beneficial to protecting the data privacy of the user, improve the data processing efficiency and the safety, and promote the smooth performance of the steel transaction process.

Preferably, the symmetric encryption calculation formula of the steel real-time transaction information in step S71 is specifically:

；

wherein ,for the encrypted symmetric encrypted ciphertext data, +.>For symmetric encryption key->For encrypted data ciphertext->For the time interval of real-time transactions +.>Trade order data for steel->Is->Sum of individual steel trade order data +.>Weight data for steel material->For the quantity data of steel material->Weight lower limit for steel quality to account for total order steel>Weight upper limit for steel quality to account for the total amount of order steel, < ->Weight integral for steel quality, +.>Any natural number used to participate in encryption.

The invention is realized byA limit operation is shown where h is the time interval of the real-time transaction. This is partly achieved by calculating the rate at which the function f (x) varies between x+h and x to obtain an approximate slope. Obtaining the rate of change of trade order data by taking the square root of the slope, which can be used to estimate the speed of fluctuation or variation of the data by +.>The sum of transaction order data x_1+x_2+x_3+ & gt x n is represented, where k is a symmetric encryption key. The data can be scaled and transformed by logarithmic operation, the complexity and confidentiality of the data can be increased in the encryption process by +. >The square sum of the weight data y and the number data z of the steel is divided by the square root of the weight range (a+b). This part is used in the encryption process to ensure that weight and quantity data are balanced within a certain range and to maintain consistency of the data by +.>Representing intervals [ a, b ]]Upper function->Is a function of the integral of (a). This section is used to introduce a comprehensive mathematical process that ensures the complexity and confidentiality of the encrypted data.

In this embodiment, a steel transaction information service platform is provided, including:

the information acquisition module is used for acquiring real-time transaction data of the first steel product of the user; acquiring real-time data of the steel by using a sensor to acquire real-time data of the multi-mode steel; the multi-mode steel real-time data comprise steel real-time video, steel price data, steel material data, steel origin data, steel specification grade data, steel mass density data and steel real-time transaction condition data;

the characteristic extraction module is used for carrying out characteristic extraction on the multi-mode steel real-time data by utilizing a characteristic engineering algorithm to generate steel transaction information characteristic data; carrying out data integration on the steel transaction information characteristic data by utilizing a knowledge graph method to generate a steel transaction information characteristic database;

The interactive view module is used for performing interactive view processing on the steel transaction information characteristic database by using an association rule learning algorithm to generate a steel characteristic real-time interactive view;

the convolution model module is used for performing expansion convolution and multi-scale sampling on the steel characteristic real-time interactable view by utilizing a convolution neural network to construct a steel characteristic convolution model;

the data binding module is used for carrying out data binding on the steel characteristic convolution model and the first steel real-time transaction data of the user by utilizing a bidirectional binding algorithm to generate a real-time steel transaction model of the user;

the data dimension reduction module is used for carrying out data dimension reduction on the user real-time steel transaction model by using a principal component analysis method to generate user second steel real-time transaction data;

the data encryption module is used for symmetrically encrypting the second steel real-time transaction data of the user by utilizing a symmetrical encryption algorithm to generate encrypted data of the steel real-time transaction of the user; and uploading the encrypted data of the user real-time steel transaction to the steel transaction information service platform by using a linear programming method, so as to realize steel transaction data management.

According to the invention, a steel transaction information service platform is established, and first steel real-time transaction data and multi-mode steel real-time data of a user are acquired through an information acquisition module and a sensor, wherein the multi-mode steel real-time data comprises steel real-time video, steel price data, steel material data, steel origin data, steel specification grade data, steel mass density data and steel real-time transaction condition data, and steel transaction data from different sources including suppliers, buyers, logistics companies and the like are integrated in a concentrated mode. The platform is beneficial to unified management and maintenance of all relevant data, improves the consistency and accuracy of the data, and can receive and process real-time steel transaction data. This enables the user to obtain up-to-date transaction information in a timely manner, including price changes, inventory conditions, etc. Real-time data processing also helps to discover potential problems or risks in advance and take corresponding measures in time, and steel transaction involves a large amount of sensitive information such as purchase amount, price, payment information and the like. The platform provides a safe data storage and transmission mechanism, and adopts measures such as encryption and identity verification to ensure the safety of data. In addition, the platform can set access control according to the roles and the authorities of the users, confidentiality and integrity of data are protected, and the platform can automatically process some conventional transaction processes, such as order generation, payment processing and the like, so that the possibility of manual intervention and errors is reduced. The processing time and cost of the transaction can be obviously reduced through an automatic flow, the efficiency and accuracy of the transaction are improved, the steel transaction information service platform has a data analysis function, a large amount of transaction data can be analyzed and mined, and tools such as real-time statistics report forms, trend analysis and the like are provided. Through deep analysis of the data, the user can obtain better business insight and make more targeted decisions.

Drawings

FIG. 1 is a flow chart of steps of a steel transaction data processing method according to the present application;

FIG. 2 is a detailed implementation step flow diagram of step S1;

FIG. 3 is a detailed implementation step flow diagram of step S2;

fig. 4 is a detailed implementation step flow diagram of step S3.

Detailed Description

It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.

The embodiment of the application provides a steel transaction information service platform and a transaction data processing method. The execution main body of the steel transaction information service platform and the transaction data processing method comprises, but is not limited to, the system: mechanical devices, data processing platforms, cloud server nodes, network uploading devices, etc. may be considered general purpose computing nodes of the present application, including but not limited to: at least one of an audio image management system, an information management system and a cloud data management system.

Referring to fig. 1 to 4, the present application provides a steel transaction data processing method, which includes the following steps:

In the embodiment of the present invention, as described with reference to fig. 1, a flow chart of steps of a steel transaction information service platform and a transaction data processing method according to the present invention is shown, where in this example, the steps of the steel transaction data processing method include:

in this embodiment, the information acquisition module is used to acquire real-time transaction data of the first steel of the user, the module needs to acquire real-time transaction data from a data source through an API, data capture or other modes, and maintain real-time performance of the data through a periodic update or push mechanism, in order to acquire real-time data of the multi-mode steel, a sensor needs to be deployed, a camera can be installed at a key position to shoot the steel so as to provide a real-time video stream, and the real-time video stream can be used for monitoring and auditing the transaction process. This may be accomplished through a wireless network, internet of things, or other suitable means of communication. The data storage system may be a cloud server or a local database for storing and managing large amounts of real-time data.

in this embodiment, the features to be extracted are determined according to the characteristics and the targets of the real-time data of the multi-mode steel. For example, the real-time data of the multi-modal steel may be cleaned and preprocessed by considering the characteristics of the steel such as size, weight, material property, and price change trend, including processing missing values, abnormal values, data conversion, and the like. The method can ensure the quality and accuracy of data, and extract the characteristic with representativeness and distinguishing degree by applying a proper characteristic extraction algorithm. Common feature extraction methods include statistical features, frequency domain features, time domain features, image features, and the like. And selecting a proper method according to specific requirements, and selecting the characteristics useful for the target task through a characteristic selection method so as to reduce the dimension and redundancy of the characteristics. Common feature selection methods include chi-square inspection, information gain, L1 regularization and the like, and a knowledge graph of steel transaction is constructed according to the domain knowledge and business requirements of the steel transaction. The knowledge graph is a structured knowledge representation mode, entities, attributes and relations are organized into a graph network so as to facilitate knowledge expression, query and reasoning, and the characteristic data of the steel transaction information and the existing knowledge graph are integrated by using a knowledge graph method. The method comprises the steps of matching and linking the extracted characteristic data with entities, attributes and relations in the knowledge graph, expanding the knowledge graph according to the requirements of the characteristic data, and adding missing entities, attributes and relations. The method can be realized by a knowledge supplementing or automatic knowledge extracting technology of a field expert, and a database structure suitable for storing the characteristic data of the steel transaction information is designed according to the characteristics and the use requirements of the characteristic data. And a proper data storage mode such as a relational database or a graph database can be selected, and the steel transaction information characteristic data subjected to characteristic extraction and knowledge graph integration can be imported into the database. The method comprises the steps of storing data in a structured mode, establishing indexes to improve the query efficiency of the data, verifying the imported data and ensuring the integrity and consistency of the data. In the subsequent operation, the characteristic data in the database are updated periodically according to the actual demand so as to maintain the real-time performance and accuracy of the data.

in the embodiment, the steel transaction information characteristic database is subjected to data preprocessing, so that the quality and accuracy of data are ensured. This includes processing missing values, outliers, duplicate data, etc. The data preprocessing can be performed by utilizing the data cleaning and data checking methods, and the association rule learning algorithm is utilized to find the association rule in the steel transaction information characteristic database. The association rule learning algorithm can help find frequent item sets and association rules in the data to reveal the association and regularity between the data, and is utilized to find association rules in the steel transaction information feature database. The association rule learning algorithm can help find frequent item sets and association rules in the data to reveal the association and regularity between the data, and according to the frequent item sets, the association rule with certain confidence and support is generated by using the association rule learning algorithm. The association rules refer to rules among association items in the data, can be expressed in the form of IF-THEN, and based on the generated association rules, real-time interactable views of steel features are created. This may be accomplished by means of a data visualization tool, a chart library, etc. to present the results of the association rules and the associated features, and the steel transaction information feature database may contain real-time data, thus requiring periodic updating of the generated interactive views. A timed task or event trigger mechanism may be set to ensure that the data presented in the view is always up to date.

in the present embodiment, data for constructing a steel characteristic convolution model is prepared. This includes steel feature data in the real-time interactive view and its corresponding tag or target variable. And ensuring the accuracy and the integrity of the data, and preprocessing the prepared data. This may include data cleansing, data normalization, feature selection, etc. Ensuring that the data meets the input requirement of the convolutional neural network, and selecting a proper convolutional neural network model structure to process the steel characteristic data. Existing pre-training models, such as VGG, resNet, or self-designed model structures, may be selected. And selecting according to actual requirements and complexity of the problems, and determining an input data format of the steel characteristic convolution model. According to the actual situation, the data in the real-time interactable view can be used as the input of the model. Typically, the data needs to be converted into tensor form for a convolution operation to determine the input data format of the steel feature convolution model. According to the actual situation, the data in the real-time interactable view can be used as the input of the model. Typically, the data needs to be converted into tensor form for convolution operations, training and tuning the constructed steel feature convolution model. This includes defining a loss function, selecting an optimizer, setting training parameters, and the like. The training data is used for back propagation and parameter updating to gradually optimize the model. And constructing a steel characteristic convolution model.

in this embodiment, the input data format of the user real-time steel transaction model is determined. According to the actual situation, the output data of the steel characteristic convolution model and the first transaction data of the user can be used as the input of the model. Typically, the data needs to be converted into a suitable format, such as tensor form, and a two-way binding algorithm is applied to bind the steel feature convolution model and the first transaction data of the user. The bidirectional binding algorithm can realize real-time interaction between the model and the data, and automatically update the output result of the model and the change of related data. Therefore, the model can adjust prediction and recommendation according to the real-time transaction data of the user, dynamic interaction between the steel transaction model and the real-time transaction data of the user is realized, and the prediction and recommendation result of the steel transaction model is adjusted according to the real-time data.

In this embodiment, a principal component analysis algorithm is applied to reduce the dimension of data of a user real-time steel transaction model. Principal component analysis is a commonly used dimension reduction technique that can convert high-dimensional data into low-dimensional data while retaining much of the information of the data. The principal component can be determined and projected by calculating the covariance matrix, eigenvalue and eigenvector of the data, and the principal component obtained by principal component analysis can select the most important feature according to the contribution degree of the principal component to the total variance. The number of the reserved main components is selected according to the requirement, so that the purpose of reducing the dimension is achieved. The principal components with higher cumulative contribution rate are usually selected to retain more data information, and the data of the user real-time steel transaction model are converted and projected by using the selected principal components. And projecting the original data to a subspace formed by the main components to obtain the real-time transaction data of the second steel product of the user. The data are representations of the original data after dimension reduction, the dimension is lower, more data information is still reserved, and the generated real-time transaction data of the second steel of the user are analyzed and applied. These data may be used for visualization, pattern discovery, statistical analysis, etc. Depending on the specific requirements, machine learning algorithms, data mining techniques, or predictive modeling, etc. may be further applied.

In this embodiment, a suitable symmetric encryption algorithm is selected for encrypting the user's second steel real-time transaction data to generate a key required by the symmetric encryption algorithm. The key is secret and is used to encrypt and decrypt data. Ensuring that the generated key is sufficiently strong and secure and taking appropriate key management measures, such as periodic rekeying, the user second steel real-time transaction data is encrypted using the selected symmetric encryption algorithm and the generated key. The data is converted into ciphertext to ensure confidentiality and security of the data. And proper filling mode and encryption mode are ensured to be used in the encryption process, so that the confidentiality and reliability of data are improved, and the encrypted user real-time steel transaction data are managed. And ensuring correct storage, backup and access control of the encrypted data so as to prevent unauthorized access and data leakage, and uploading the encrypted data of the user real-time steel transaction to the steel transaction information service platform by using a linear programming method. Linear programming is a commonly used optimization method that finds the best solution based on specific linear constraints and objective functions. In this case, the linear programming may be used to ensure the integrity and accuracy of the uploaded data, and the user real-time steel transaction encrypted data is uploaded to the steel transaction information service platform using the linear programming method. Linear programming is a commonly used optimization method that finds the best solution based on specific linear constraints and objective functions. In this case, linear programming may be used to ensure the integrity and accuracy of the uploaded data.

In this embodiment, a detailed implementation step flow diagram of the step S1 is described with reference to fig. 2, and in this embodiment, the detailed implementation step of the step S1 includes:

In this embodiment, an information acquisition module is set and configured to ensure that the information acquisition module is connected with a data source of a first steel system of a user, is connected to the first steel system of the user, acquires access rights of real-time transaction data, initiates a data request through the information acquisition module, acquires user purchase data, user transaction amount data, user identity information data, a user transaction timestamp and user transaction status data, stores or transmits the acquired real-time transaction data for subsequent processing and analysis, acquires real-time image data of the steel through a video sensor, processes and encodes the acquired image data, generates real-time video of the steel, activates an ultrasonic sensor device to enable the ultrasonic sensor device to perform texture detection on the steel, transmits an ultrasonic signal to the steel and receives a returned signal, analyzes the texture data of the steel through characteristics of the signals, processes and analyzes the acquired texture data, extracts texture data, steel specification grade data and steel quality density data of the steel, is connected to a steel transaction information service platform, ensures that the access rights are provided, initiates the request and acquires real-time transaction condition data of the steel, includes information such as market supply condition, transaction condition data and the accuracy and availability of the acquired data are ensured.

In this embodiment, a detailed implementation step flow diagram of the step S2 is described with reference to fig. 3, and in this embodiment, the detailed implementation step of the step S2 includes:

In this embodiment, a suitable feature point detection algorithm, such as SIFT (scale invariant feature transform), SURF (speeded up robust feature) or ORB (Oriented FAST and Rotated BRIEF), is used to detect feature points of steel real-time data of each mode, extract key information such as position, scale and direction of each feature point, generate steel transaction information feature codes, process and convert the coded data by using a feature engineering method, including dimension reduction, normalization, selection of important features, etc., perform feature extraction, extract features related to steel transaction information, such as transaction amount, transaction timestamp, transaction state, generate a steel transaction information feature data set containing the feature extracted steel transaction information features, map the steel transaction information feature data to a knowledge graph model, each feature is regarded as a node or attribute, relationships among the nodes are established, potential concept nodes in steel transaction information feature data are found by utilizing a knowledge graph reasoning and mining algorithm, feature extraction and description are carried out on each concept node, steel transaction information concept node data are generated, steel transaction information concept node data and corresponding relationships thereof are combined, relationship connection among the nodes is established, node relationship inference and combination are carried out according to association rules, attribute similarity and the like among the concept nodes, a steel transaction information node relationship data set is generated, the data set contains complete node relationship information, the structure and connection of steel transaction information are reflected, information in the node relationship data set is imported into a feature database, and reasonable storage and index of the data are ensured, and (3) cleaning and verifying the data of the characteristic database, ensuring the quality and consistency of the data, and completing the construction and integration of the characteristic database, wherein the characteristic database contains the characteristic data and the node relation of the multi-mode steel transaction information, and can support the subsequent data analysis and application.

In this embodiment, the specific steps of step S23 are as follows:

In this embodiment, concept attribute labeling is performed on concept node data of steel transaction information, corresponding attributes are labeled for each concept node, and concept nodes and corresponding attribute values are combined to generate a steel transaction information concept attribute data set. And ensuring the accuracy and the integrity of the data set, and formulating a set of concept relation rules according to the knowledge and experience in the steel transaction field. The rules may be based on semantic associations, co-occurrence relationships, or other logical relationships between concept nodes, using prepared tools and rule libraries to conduct concept relationship mining on steel transaction information concept node data. And (3) applying rules to find out the association relation between the concept nodes, recording the relation between the concept nodes according to the mining result, generating a steel transaction information concept relation data set, and combining the relations with the same nodes to form a node relation set. This may be accomplished by combining or aggregating the relationships of the same nodes, recording a set of node relationships, and generating a final steel transaction information node relationship dataset. Ensuring consistency and availability of the data sets.

In this embodiment, as described with reference to fig. 4, a detailed implementation step flow diagram of the step S3 is described, and in this embodiment, the detailed implementation step of the step S3 includes:

In this embodiment, preprocessing is performed on the feature database, including operations such as data cleaning, deduplication, and standardization, so as to ensure quality and consistency of data, and an appropriate association rule learning algorithm, such as an Apriori algorithm or an FP-Growth algorithm, is selected. The algorithms can mine frequent item sets from the characteristic database, apply the association rule learning algorithm, and mine the frequent item sets from the steel transaction information characteristic database. The frequent item sets are feature combinations frequently occurring in transaction data, and each transaction is converted into a feature vector according to the mining result of the frequent item sets. The feature vector may represent feature combinations and their corresponding frequency in the transaction, and according to the visualization target and data characteristics, a suitable visualization tool is selected, such as Python Matplotlib, seaborn or Tableau, and the like, and the feature vector data is processed and converted according to the visualization requirement, so as to construct a visualization view. For example, data clustering, dimension reduction, or other related data processing operations may be performed to design appropriate visual view forms, such as bar graphs, scatter graphs, thermodynamic diagrams, etc., based on the goals and data characteristics. The method comprises the steps of ensuring that views can clearly present characteristic conditions of steel transaction information, generating a visual view of the characteristics of the steel transaction information according to designed characteristic vector data and visual view forms by using selected visual tools, compiling codes for the characteristic visual view by using a selected JavaScript library, and adding corresponding interaction functions. According to the documents and examples of the library, the interactive effect is realized by using an appropriate API and method, the correctness and stability of the interactive function are tested, the code and the user experience are optimized to improve the effect of the visual view of the interactive feature, the visual view implementing the interactive function is embedded into the webpage, and the real-time interactive view of the steel feature is generated. Ensuring that the interactive functionality is able to provide a user-friendly data exploration and analysis experience.

In this embodiment, the specific steps of step S4 are as follows:

In this embodiment, the view data is subjected to convolution preprocessing by using a convolution neural network, so that the data with a time sequence relationship can be processed, and spatial characteristics are reserved, the view data is input into the convolution neural network, the characteristic representation in the view is extracted and learned, a steel transaction information characteristic sample set is generated, the steel transaction information characteristic sample set is subjected to super-pixel segmentation, and is divided into continuous areas with similar characteristics, the original data set is divided into a group of mutually connected areas by using a super-pixel algorithm, each area is called a super-pixel, convolution operation is applied to each super-pixel, a low-dimensional convolution characteristic sequence is extracted, the low-dimensional convolution characteristic sequence is a characteristic representation of the super-pixel area, the dimension of the data is reduced, the spatial information of the area is reserved, and edge characteristic enhancement processing is performed on the low-dimensional convolution characteristic sequence by using an expansion convolution (Dilated Convolution) algorithm, the expansion convolution increases the receptive field by introducing holes (conditions) into the convolution kernel, so that the convolution operation can perceive wider local areas, the edge characteristic strengthening treatment can improve the perception capability of a model on the characteristic edges of steel transaction data, key information can be better captured, after the expansion convolution is carried out on a low-dimensional convolution characteristic sequence, a steel transaction information convolution characteristic network is obtained, the characteristic representation capability is stronger, a multi-scale sampling algorithm is used for carrying out spatial pyramid pooling multi-layer sampling on the steel transaction information convolution characteristic network, the multi-scale sampling can capture characteristic information under different scales by carrying out pooling operation on characteristic diagrams on different scales, the spatial pyramid pooling divides the characteristic diagrams into areas with different scales, and pooling operation is carried out on each area, the method comprises the steps of obtaining a characteristic representation with fixed length, carrying out pooling on a spatial pyramid to divide the characteristic graph into areas with different scales, carrying out pooling operation on each area to obtain the characteristic representation with fixed length, enabling a combined classifier to integrate results of different classifiers, improving prediction accuracy and robustness, carrying out weighted combination on the results of the combined classifier by using a weighted comprehensive calculation formula, carrying out modeling on the basis of a data mining algorithm of the combined classifier algorithm, finding out associated modes and rules from a steel transaction information convolution characteristic graph, and carrying out prediction and decision to finally generate a steel transaction information convolution characteristic model, and carrying out classification and prediction on new data.

In this embodiment, the combined classifier weighted comprehensive calculation formula in step S45 is specifically:

；

wherein ,for combining classifier weight coefficients, +.>Is->Personal classifier, < >>For the number of basis classifiers, +.>Is the firstWeights of the individual basis classifier in the combined classifier,/->Is->Predictive value of weight by the individual basis classifier, < +.>For inputting sample values of the initial basis classifier, < +.>Sum of predicted results for the result values for the basis classifier,/->Is->Predictive outcome of the outcome value by the personal classifier, < >>Is->Accuracy of the prediction of the radix classifier, +.>For the number of classification results, +.>Is->The personal classifier pair->Weights of the individual basis classifier, +.>Is->Classifying results of the individual basis classifier on the sample, < >>For sample->In->The value of the individual basis classifier.

The invention is realized byRepresenting the sum of the base classifier weight coefficients multiplied by the predicted value of each base classifier pair weight. By weighted summing the predicted values of the individual base classifiers, the degree of contribution of each base classifier to the final combined classifier can be captured. This is beneficial for taking into account the predictive power of the different base classifiers in combination. By passing throughMultiplying the number of predicted results and classified results representing the result value by the base classifier The products are multiplied and square root is divided. It may consider a balanced relationship between the number of classification results and the predicted results of the base classifier on the result values. This is beneficial in adjusting the final weight according to the number of classification results and the predictive accuracy of each classification result. By->The Sigmoid function maps the predicted values of the base classifier weights, converting the predicted values into probability values ranging from 0 to 1. This is beneficial to normalize the predictors so that they can represent the degree of weight impact of the individual basis classifiers and can be interpreted and understood by means of probability values. By->Representing the classification result of the base classifier on the samples multiplied by the weights between the base classifiers. This may allow for further adjustment of the final weights taking into account interactions and dependencies between different base classifiers. The formula calculates and adjusts the weight coefficient of the combined classifier by comprehensively considering the factors such as weight prediction, result prediction, accuracy, quantity and sample classification result of the base classifier. This is beneficial to more fully evaluating the contribution of the base classifier, balancing the different factors, and further improving the performance and accuracy of the combined classifier.

In this embodiment, the specific steps of step S5 are as follows:

In this embodiment, the selected bi-directional binding library or framework is imported and an empty steel transaction bi-directional binding framework is created, defining the data items to be bound, such as steel characteristics, transaction date, transaction price, etc., in the steel characteristics convolution model. The data items are supposed to correspond to elements in the user interface, the data in the steel characteristic convolution model and the steel transaction bidirectional binding frame are bound by using a bidirectional binding algorithm, synchronous updating of the data is ensured, and the data items in the steel characteristic convolution model and the first transaction data of the user are bound by using the steel transaction bidirectional binding frame. The data monitor can monitor the change of the transaction data in the user interface, when a user carries out new steel transaction, the data monitor can capture the change of the transaction data, and the data monitor transmits the captured data change to the steel transaction bidirectional binding frame so as to update the data in the model and the user interface, and the updated user steel transaction data can be used for generating a user real-time steel transaction model so as to carry out real-time monitoring and analysis.

In this embodiment, the specific steps of step S6 are as follows:

In this embodiment, the data is standardized, which may be converted into standardized data with the same scale and unit by using methods such as mean normalization and standard deviation normalization, a Principal Component Analysis (PCA) algorithm is applied, the real-time steel transaction data with high dimensions is reduced to lower dimensions, in the PCA, a covariance Matrix of the data is calculated, and feature value decomposition is performed to obtain feature values and feature vectors, the feature vectors corresponding to the previous N feature values are selected as principal components according to the size ordering of the feature values, the selection of N may be determined according to the actual requirements and the interpretation capability of the data, the standardized data is divided into a plurality of sub-matrices by using an adaptive dividing method based on the features of the real-time steel transaction data, the adaptive dividing method is a method for dynamically dividing the standardized data according to the features of the data, the divided sub-matrices may be selected according to different requirements, such as time, geographic location, steel type, etc., the user real-time steel transaction data is subjected to Non-negative Matrix decomposition (N-negative Matrix, nmfactor Matrix is an optimal Matrix, NMF is a Non-optimal Matrix, and the Non-real-time transaction data is represented by a Non-optimal Matrix, and the Non-optimal Matrix is obtained by the method, and the Non-optimal Matrix is an optimal Matrix, the Non-real-time transaction data is represented by the Non-optimal Matrix; the other is a weight matrix, which describes the weight of each principal component in the principal component matrix, calculates the covariance matrix according to the principal component matrix of the user real-time steel transaction, calculates the eigenvalue of each principal component and the corresponding eigenvector by the covariance matrix, sorts the principal eigenvalue and the corresponding eigenvector according to the eigenvalue, and reconstructs the user second steel real-time transaction data containing the characteristics with more information and importance by using the selected eigenvalue and eigenvector.

In this embodiment, the specific steps of step S7 are as follows:

In this embodiment, for the second steel real-time transaction data of the user, the selected symmetric encryption algorithm is used to encrypt and decrypt the data, the symmetric encryption algorithm uses the same secret key to encrypt and decrypt the data, the encryption calculation formula is determined according to the steel real-time transaction information and the selected symmetric encryption algorithm, the encryption calculation formula is used to encrypt the symmetric data ciphertext of the steel real-time transaction to obtain the symmetric encryption data of the steel real-time transaction, the encryption calculation formula can be based on some attributes of the transaction information or the customized encryption rule to ensure that the generated encryption data has safety and irreversibility, the linear programming method is used to perform network scheduling slicing on the symmetric encryption data of the steel real-time transaction, the linear programming method can consider various limiting conditions and requirements of the transaction data, split the data into a plurality of parts or perform optimal scheduling to meet the requirements of network transmission, the scheduling slicing can be optimized based on factors such as network bandwidth, transmission speed, data size and the like to improve the efficiency and stability of data transmission, the generated symmetric data of the steel real-time transaction is uploaded to the steel transaction information service platform, the uploaded data is ensured to have proper safety and security measures are provided, the transaction information can be prevented from being leaked, and the transaction data can be analyzed and stored, and the transaction information can be analyzed.

In this embodiment, the symmetric encryption calculation formula of the steel real-time transaction information in step S72 is specifically:

；/>

The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

It will be understood that, although the terms "first," "second," etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.

The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A steel transaction data processing method, comprising the steps of:

2. The method according to claim 1, wherein the specific steps of step S1 are:

3. The method according to claim 1, wherein the specific steps of step S2 are:

4. A method according to claim 3, wherein the specific step of step S24 is:

5. The method according to claim 1, wherein the specific step of step S3 is:

6. The method according to claim 1, wherein the specific step of step S4 is:

the combined classifier weighting comprehensive calculation formula in step S45 specifically includes:

；

wherein ,for combining classifier weight coefficients, +.>Is->Personal classifier, < >>For the number of basis classifiers, +.>Is->Weights of the individual basis classifier in the combined classifier,/->Is->Predictive value of weight by the individual basis classifier, < +.>For inputting sample values of the initial basis classifier, < +.>Sum of predicted results for the result values for the basis classifier,/->Is->Predictive outcome of the outcome value by the personal classifier, < >>Is->Accuracy of the prediction of the radix classifier, +.>For the number of classification results, +.>Is->The personal classifier pair->Weights of the individual basis classifier, +.>Is->Classifying results of the individual basis classifier on the sample, < >>For sample->In->The value of the individual basis classifier.

7. The method according to claim 1, wherein the specific step of step S5 is:

8. The method according to claim 1, wherein the specific step of step S6 is:

9. The method according to claim 1, wherein the specific step of step S7 is:

step S74: uploading the encrypted steel transaction data of the user in real time to a steel transaction information service platform to realize steel transaction data management;

the symmetric encryption calculation formula of the steel real-time transaction information in the step S72 specifically comprises the following steps:

；

wherein ,for symmetrical encryption after encryptionCiphertext data->For symmetric encryption key->For encrypted data ciphertext- >For the time interval of real-time transactions +.>Trade order data for steel->Is->Sum of individual steel trade order data +.>Weight data for steel material->For the quantity data of steel material->Weight lower limit for steel quality to account for total order steel>Weight upper limit for steel quality to account for the total amount of order steel, < ->Weight integral for steel quality, +.>Any natural number used to participate in encryption.

10. A steel transaction information service platform, comprising: