CN116976894A - Artificial intelligence electronic commerce data protection method and system - Google Patents

Artificial intelligence electronic commerce data protection method and system Download PDF

Info

Publication number
CN116976894A
CN116976894A CN202311017169.3A CN202311017169A CN116976894A CN 116976894 A CN116976894 A CN 116976894A CN 202311017169 A CN202311017169 A CN 202311017169A CN 116976894 A CN116976894 A CN 116976894A
Authority
CN
China
Prior art keywords
data
commerce data
electronic commerce
abnormal
protection
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
CN202311017169.3A
Other languages
Chinese (zh)
Inventor
连少聪
潘佳
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chengdu Xijie Technology Co ltd
Original Assignee
Chengdu Xijie Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chengdu Xijie Technology Co ltd filed Critical Chengdu Xijie Technology Co ltd
Priority to CN202311017169.3A priority Critical patent/CN116976894A/en
Publication of CN116976894A publication Critical patent/CN116976894A/en
Withdrawn legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q20/00Payment architectures, schemes or protocols
    • G06Q20/38Payment protocols; Details thereof
    • G06Q20/382Payment protocols; Details thereof insuring higher security of transaction
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/21Design, administration or maintenance of databases
    • G06F16/215Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/23Updating
    • G06F16/2308Concurrency control
    • G06F16/2315Optimistic concurrency control
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/28Databases characterised by their database models, e.g. relational or object models
    • G06F16/284Relational databases
    • G06F16/285Clustering or classification
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/10Pre-processing; Data cleansing
    • G06F18/15Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/24Classification techniques
    • G06F18/243Classification techniques relating to the number of classes
    • G06F18/2433Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60Protecting data
    • G06F21/602Providing cryptographic facilities or services
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L9/00Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
    • H04L9/008Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols involving homomorphic encryption

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Computer Security & Cryptography (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Accounting & Taxation (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • General Business, Economics & Management (AREA)
  • Probability & Statistics with Applications (AREA)
  • Software Systems (AREA)
  • Computer Hardware Design (AREA)
  • General Health & Medical Sciences (AREA)
  • Bioethics (AREA)
  • Health & Medical Sciences (AREA)
  • Strategic Management (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Quality & Reliability (AREA)
  • Finance (AREA)
  • Storage Device Security (AREA)

Abstract

The invention relates to the technical field of data security, in particular to an artificial intelligence electronic commerce data protection method and system. The method comprises the following steps: carrying out data preprocessing on the original electronic commerce data to generate standard electronic commerce data; performing electronic commerce data anomaly detection processing on standard electronic commerce data to generate electronic commerce data anomaly values; data division is carried out on the standard electronic commerce data according to the electronic commerce data abnormal value, and abnormal electronic commerce data and conventional electronic commerce data are respectively generated; performing abnormal field extraction processing on the abnormal electronic commerce data to generate abnormal field data; e-business protection decision design is carried out according to the abnormal field data, and E-business protection decisions are generated; e-commerce data protection processing is carried out on conventional E-commerce data by utilizing an E-commerce protection decision, and safe E-commerce data is generated; and carrying out homomorphic encryption processing on the secure electronic commerce data, and carrying out real-time updating to generate the real-time encrypted secure electronic commerce data. According to the invention, protection decisions are designed through the abnormal data, so that the safety of the E-commerce data is ensured.

Description

Artificial intelligence electronic commerce data protection method and system
Technical Field
The invention relates to the technical field of data security, in particular to an artificial intelligence electronic commerce data protection method and system.
Background
In the e-commerce field, a large amount of user data, transaction information and business secrets are widely used, and a large amount of sensitive information such as payment account numbers, addresses, contact ways and the like are stored and transmitted, and meanwhile, the interest of hackers and malicious attackers is attracted. In order to prevent data leakage, information theft or unauthorized access, an effective electronic commerce data protection method becomes an indispensable security measure. However, the traditional e-commerce data protection method is complicated when solving increasingly complex attack means, and is not easy to find the hidden attack means, so that the transaction safety of electronic commerce cannot be guaranteed.
Disclosure of Invention
Based on the above, the present invention provides an artificial intelligence e-commerce data protection method and system, so as to solve at least one of the above technical problems.
To achieve the above object, an e-commerce data protection method of artificial intelligence, the method comprises the following steps:
step S1: acquiring original E-commerce data; carrying out data preprocessing on the original electronic commerce data to generate standard electronic commerce data;
step S2: performing electronic commerce data anomaly detection processing on standard electronic commerce data to generate electronic commerce data anomaly values; data division is carried out on the standard electronic commerce data according to the electronic commerce data abnormal value, and abnormal electronic commerce data and conventional electronic commerce data are respectively generated;
Step S3: performing abnormal field extraction processing on the abnormal electronic commerce data to generate abnormal field data; e-business protection decision design is carried out according to the abnormal field data, and E-business protection decisions are generated; e-commerce data protection processing is carried out on conventional E-commerce data by utilizing an E-commerce protection decision, and safe E-commerce data is generated;
step S4: and carrying out homomorphic encryption processing on the secure electronic commerce data by utilizing homomorphic encryption technology, and carrying out real-time updating to generate real-time encrypted secure electronic commerce data.
In the E-commerce field, the original data are usually from a plurality of different data sources, and the problems of non-uniform data format, deletion, errors and the like possibly exist, and the original E-commerce data can be cleaned, the deletion value is filled and the errors are corrected through data preprocessing, so that the standard E-commerce data are generated, the data quality is improved, the errors in subsequent processing are reduced, and the effectiveness and the accuracy of the subsequent steps are ensured. The data abnormality of the standard electronic commerce data is possibly caused by unexpected errors, malicious attacks or system faults, potential abnormal data can be found in time through electronic commerce data abnormality detection processing, abnormal electronic commerce data and conventional electronic commerce data are further separated, special processing and analysis are conducted on the abnormal data, interference of the abnormal data on the conventional data is reduced, and data processing efficiency is improved. The extraction of the abnormal field data can help reveal specific reasons for causing the abnormality, so that potential data security threat can be better known, and a targeted protection strategy and rule can be formulated according to the abnormal field data through the design of an e-commerce protection decision so as to cope with various abnormal conditions, improve the security of an e-commerce system and prevent potential risks of data leakage, tampering, theft and the like. The homomorphic encryption technology is an advanced encryption method, which allows data operation to be performed in an encryption state, and results are still kept encrypted, so that the technology enables the data to keep the encryption state in the transmission and processing processes, the confidentiality of the data can be protected even if the data is attacked, the security of the data is ensured to be continuously maintained when the data is updated in real time, security measures can be timely applied even if the data changes, and the homomorphic encryption can effectively protect the security of e-commerce data in the transmission and processing processes by combining the operation of the real-time updating. Therefore, the artificial intelligence electronic commerce data protection method can be used for analyzing historical attack means to relate to protection decisions when the electronic commerce data protection method is used for complex attack means, so that abnormal behaviors are analyzed and attack trends are predicted, and hidden attack means can be identified, so that the transaction safety of electronic commerce is ensured.
Preferably, step S1 comprises the steps of:
step S11: acquiring original E-commerce data;
step S12: e-commerce data effective time sequence screening is carried out on original E-commerce data, and effective E-commerce data is generated;
step S13: data screening is carried out on the effective electronic commerce data according to a preset protection data target, and target electronic commerce data is generated;
step S14: performing data cleaning processing on the target electronic commerce data to generate cleaning electronic commerce data;
step S15: and carrying out data standardization processing on the cleaning electronic commerce data by using a minimum-maximum standardization method to generate standard electronic commerce data.
The original electronic commerce data may come from different data sources, such as transaction records, user behaviors, commodity information and the like, and a data basis is provided for the subsequent method. In an e-commerce environment, the effective time range of data is usually limited, and by carrying out effective time sequence screening on original e-commerce data, outdated, invalid or no longer needed data can be eliminated, so that the scale and complexity of a data set are reduced, the efficiency of subsequent processing is improved, the storage space is saved, and the timeliness of the data is ensured. The data subjected to effective time sequence screening is further screened according to a preset protection data target, such as protecting user privacy, preventing fraud, monitoring abnormal transactions and the like, focusing on important parts needing special protection in the e-commerce data, reducing the processing of irrelevant data and optimizing the data processing flow. The original data may contain problems such as noise, errors, missing values and the like, and the problems may cause errors and inaccuracy in the subsequent data processing process, and the errors in the data are corrected through the data cleaning process, so that the missing values are filled, the noise is eliminated, the quality and consistency of the data are ensured, and the reliability and accuracy of the subsequent processing steps are improved. The scales and ranges of different data may be different, which makes it difficult to effectively compare and analyze the data, and the cleaned e-commerce data is subjected to standardized processing by a min-max standardization method, so that the data is in a uniform scale range, and subsequent analysis and processing are facilitated, which helps to ensure comparability among different data features, and meanwhile, improves the efficiency of data processing and analysis.
Preferably, step S2 comprises the steps of:
step S21: carrying out electronic commerce data anomaly detection processing on standard electronic commerce data by utilizing an electronic commerce data anomaly detection algorithm to generate electronic commerce data anomaly values;
step S22: e-commerce data abnormal value is judged according to a preset E-commerce data abnormal threshold value, and when the E-commerce data abnormal value is larger than the E-commerce data abnormal threshold value, standard E-commerce data corresponding to the E-commerce data abnormal value is marked as abnormal E-commerce data; and when the abnormal value of the electronic commerce data is not larger than the abnormal threshold value of the electronic commerce data, marking the standard electronic commerce data corresponding to the abnormal value of the electronic commerce data as conventional electronic commerce data.
The invention processes the standard electronic commerce data by utilizing the electronic commerce data anomaly detection algorithm to identify the possible anomaly value, and the system can discover the possible fraudulent transaction, the possible anomaly behavior or other potential risk conditions by carrying out anomaly detection on the standard electronic commerce data, which is helpful for discovering and identifying the anomaly condition in an early stage and provides important basis for subsequent data processing and protection decision. Judging an abnormal value of the electronic commerce data according to a preset electronic commerce data abnormal threshold value, and marking the corresponding standard electronic commerce data as abnormal electronic commerce data if the abnormal value of the electronic commerce data is larger than the electronic commerce data abnormal threshold value; in contrast, if the e-commerce data outlier is not greater than the e-commerce data outlier threshold, the corresponding standard e-commerce data is marked as regular e-commerce data. Through the judging and marking process, the system can effectively distinguish abnormal and normal E-commerce data, thereby providing important guidance for subsequent protection decisions.
Preferably, the e-commerce data anomaly detection algorithm in step S21 is as follows:
wherein K is expressed as an electronic commerce data abnormal value, N is expressed as a data category of standard electronic commerce data, y i An e-commerce data mean denoted as class i,the historical average value of the e-commerce data expressed as the i-th class is expressed as the comment similarity of the standard e-commerce data, V is expressed as the repetition degree of purchasing users of the standard e-commerce data, u is expressed as the transaction abnormal weight information of the standard e-commerce data, V is expressed as the order quantity of the standard e-commerce data, t is expressed as the time length of accessing commodities involved in accessing the standard e-commerce data, and delta is expressed as the abnormal adjustment value of the abnormal value of the e-commerce data.
The invention utilizes an electronic commerce data anomaly detection algorithm which comprehensively considers the data class N and the electronic commerce data mean value y of the ith class of standard electronic commerce data i Class i e-commerce data history meansThe method comprises the steps of commentary similarity o of standard electronic commerce data, repetition V of purchasing users of the standard electronic commerce data, abnormal transaction weight information u of the standard electronic commerce data, order quantity V of the standard electronic commerce data, commodity accessing time length t related to accessing the standard electronic commerce data and interaction relation among functions to form a functional relation:
That is to say,the data category of the standard e-commerce data represents the division of the e-commerce data on different categories, such as different commodity categories or different user behavior types; the average value of the e-commerce data of the i-th class is expressed as the average value of the standard e-commerce data on the class; the historical average of the e-commerce data of the i-th class represents the average of the data class in the historical record and is used for comparing with the current standard e-commerce data average; the standard electronic commerce data comment similarity reflects the similarity degree of comment information in the standard electronic commerce data and can be used for evaluating the consensus degree of users on goods or services; the purchasing user repetition of the standard electronic commerce data is the purchasing repetition frequency of the user purchasing the same commodity in the standard electronic commerce data; standard business e-business data exchangeThe easy-to-abnormality weight information is used for giving weights with different abnormality degrees to the electronic commerce data, and is related to different types of transaction abnormalities; the order quantity of the standard e-commerce data reflects the transaction activity of the data class; the length of time for accessing the goods involved in accessing the standard electronic commerce data represents the time range of goods or services involved in accessing the standard electronic commerce data by the user, possibly reflecting the frequency of consumption behavior of the user. The function relation calculates abnormal values of the electronic commerce data through the multi-dimensional parameters, achieves comprehensive and personalized abnormal detection, provides a more accurate and effective analysis tool for electronic commerce data protection, helps an electronic commerce platform to find potential abnormal behaviors in time, and ensures data safety and business robust development. And the function relation is adjusted and corrected by utilizing the abnormal adjustment value delta of the abnormal value of the electronic commerce data, so that the error influence caused by abnormal data or error items is reduced, the abnormal value K of the electronic commerce data is more accurately generated, and the accuracy and the reliability of the electronic commerce data abnormal detection processing of the standard electronic commerce data are improved. Meanwhile, the weight information and the adjustment value in the formula can be adjusted according to actual conditions and are applied to different standard E-commerce data, so that the flexibility and applicability of the algorithm are improved.
Preferably, step S3 comprises the steps of:
step S31: carrying out characteristic extraction on the abnormal electronic commerce data to generate abnormal electronic commerce characteristic data;
step S32: carrying out log file tracking of the abnormal E-commerce characteristic data on the abnormal E-commerce characteristic data to generate an abnormal log file;
step S33: carrying out extraction processing on the abnormal field of the log file according to the abnormal log file to generate abnormal field data;
step S34: carrying out attack type identification of E-commerce data on the abnormal field data to generate attack type data;
step S35: performing exception access user field extraction on the exception field data to generate an exception access user field;
step S36: e-business protection decision design is carried out according to the attack type data and the abnormal access user field, and E-business protection decisions are generated;
step S37: and carrying out electronic commerce data protection processing on the conventional electronic commerce data by utilizing the electronic commerce protection decision to generate safe electronic commerce data.
The method and the device perform feature extraction on the data marked as the abnormal electronic commerce data, extract meaningful feature information such as transaction amount, transaction time, geographical position and the like from the abnormal electronic commerce data through the feature extraction, and the feature information is helpful for subsequent log file tracking and abnormal field extraction, so that a foundation for performing deep analysis on the abnormal data is provided. The abnormal E-commerce characteristic data is tracked by the log file, and the change and trend of the abnormal data are recorded in real time by writing the abnormal E-commerce characteristic data into the log file, so that the abnormal situation can be monitored in real time, and new abnormal modes and behaviors can be found in time. According to the method, the system can find the abnormal field, namely key information related to abnormal behaviors, such as abnormal transaction types, abnormal user IDs, abnormal IP addresses and the like, by analyzing the content in the log file, so that the system is helpful for further understanding abnormal conditions and provides detailed data support for subsequent attack type identification and E-commerce protection decision. The abnormal field data is analyzed to identify the attack type of the e-commerce data, the attack type identification can be performed based on technologies such as machine learning, deep learning and the like, different types of attack behaviors such as fraud, network attack and the like are identified through a training model, the source and the property of the abnormal data can be quickly and accurately determined, and a basis is provided for making a targeted e-commerce protection decision. The method comprises the steps of extracting an abnormal access user field from abnormal field data, and identifying an abnormal access user, namely a user suspected of participating in abnormal behaviors by analyzing user information in the abnormal field data, so that the method is beneficial to further tracking and monitoring sources of the abnormal behaviors and helping an e-commerce platform to take necessary protective measures. E-commerce protection decision design is carried out according to attack type data and abnormal access user fields, corresponding protection strategies are formulated according to the identified attack types and abnormal access users, for example, a risk scoring mechanism or transaction limits can be set for fraudulent transaction behaviors, and blocking treatment can be carried out for malicious attackers, so that targeted protection measures can be adopted for different abnormal conditions, and the safety and reliability of E-commerce data are improved. And the regular electronic commerce data is protected by utilizing an electronic commerce protection decision to generate safe electronic commerce data, and the regular electronic commerce data is safely processed according to a formulated protection strategy, so that the data is not influenced by potential abnormal behaviors, and the normal operation of an electronic commerce platform and the safety of user data are ensured.
Preferably, step S31 comprises the steps of:
step S311: carrying out e-commerce class data average value calculation on the conventional e-commerce data by using average value calculation, and generating classified conventional e-commerce average value data;
step S312: e-commerce data safety interval design is carried out according to classified conventional E-commerce mean value data, and a conventional E-commerce data interval is generated;
step S313: and carrying out abnormal electronic commerce data extraction on the abnormal electronic commerce data based on the conventional electronic commerce data interval to generate abnormal electronic commerce characteristic data.
The invention processes the conventional electronic commerce data by using a mean value calculation method to generate a mean value of electronic commerce type data, wherein the electronic commerce type data refer to type data with similar characteristics, such as sales of different commodity types, the number of registered sources of users and the like, and the system can obtain typical performance of the type data by calculating the mean value of each electronic commerce type data and serve as a reference standard for subsequent anomaly detection. And designing a safety interval of the e-commerce data according to the classified conventional e-commerce mean value data, wherein the safety interval of the e-commerce data is a range, which represents the allowable variation range of the classified data, and the normal fluctuation range of the classified data is defined by setting the safety interval, so that the data exceeding the range may belong to abnormal conditions. And extracting abnormal electronic commerce data based on a conventional electronic commerce data interval, identifying data exceeding the electronic commerce data safety interval as abnormal electronic commerce data, and generating corresponding abnormal characteristic data which contain information exceeding a normal range, so that the subsequent abnormal detection and protection decision are facilitated.
Preferably, step S36 comprises the steps of:
step S361: designing a user authenticity judging mechanism according to the abnormal access user field, and generating a user authenticity judging mechanism;
step S362: optimizing a user authenticity judging mechanism by using a user authenticity screening algorithm to generate an optimized user authenticity judging mechanism;
step S363: carrying out protection scheme design according to the attack type data to generate attack protection scheme data;
step S364: e-business protection decision integration is carried out according to the optimized user authenticity judgment mechanism and the protection scheme, and E-business protection decision is generated.
The invention designs a user authenticity judging mechanism according to the abnormal access user field, wherein the user authenticity judging mechanism aims at verifying the real identity of the user so as to identify the user with a potential malicious attacker or false identity, and the mechanism can comprise various verification means, such as short message verification codes, face recognition, voiceprint recognition and the like, so as to improve the reliability and the safety of user identity verification. The user authenticity judging mechanism is optimized by utilizing a user authenticity screening algorithm, the accuracy of judging the user authenticity is improved by analyzing historical data and behaviors, and the optimized user authenticity judging mechanism can better distinguish the real user from false identities and reduce fraudulent behaviors and attack behaviors. The design of the protection scheme is carried out according to attack type data, the attack type data provides detailed information about different attack behaviors, such as fraudulent transactions, malicious scripts and the like, and corresponding protection strategies, such as transaction risk scores, abnormal transaction detection, user behavior analysis and the like, are formulated according to the information so as to cope with the different attack behaviors. Integrating the optimized user authenticity judging mechanism with a protection scheme to generate an e-commerce protection decision, wherein the e-commerce protection decision comprehensively considers verification of user authenticity and different types of attack protection strategies to form a set of complete protection measures, and the decisions are applied to subsequent conventional e-commerce data protection processing to ensure the safety and reliability of e-commerce data.
Preferably, the user authenticity screening algorithm in step S362 is as follows:
p represents the optimized user authenticity score, a represents the user e-commerce platform credit rating, r represents the network address anomaly weight information generated according to the network address accessing the e-commerce data, b represents the initial user authenticity score generated according to the user authenticity judging mechanism, m represents the proxy anomaly weight information generated according to the user agent accessing the e-commerce data, c represents the user e-commerce data access frequency, epsilon represents the user liveness score, d represents the user anomaly behavior score, tau represents the anomaly adjustment value optimizing the user authenticity score.
The invention utilizes a user authenticity screening algorithm which comprehensively considers the interaction relation among a user E-commerce platform credit rating a, network address abnormal weight information r generated according to a network address accessing E-commerce data, an initial user authenticity score b generated according to a user authenticity judging mechanism, agent abnormal weight information m generated according to a user agent accessing E-commerce data, a user E-commerce data access frequency c, a user liveness score epsilon, a user abnormal behavior score d and functions to form a functional relation:
That is to say,the credit rating of the user e-commerce platform represents the credit rating of the user on the e-commerce platform, which is an important index and reflects whether the user has bad records or credit problems in the past behaviors; the network address anomaly weight information generated according to the network address accessing the e-commerce data is expressed as being related to the trust degree of different network addresses used by the user; the initial user authenticity score generated according to the user authenticity judging mechanism reflects the user authenticity preliminary judging result; according to accessing electronic commerce dataThe agent anomaly weight information generated by the user agent is used for reflecting the credibility of the user agent; the data access frequency of the user electronic commerce is expressed as the activity of the user and is related to the use habit and behavior of the user; the user liveness score is expressed as being obtained based on the access time length and the access frequency of the user on the e-commerce platform; the user abnormal behavior score is used to reflect whether the user has abnormal or suspicious behavior. The behavior characteristics (such as access frequency, abnormal behavior and the like) of the functional relation type user are comprehensively considered with information such as credit rating and the like, so that the authenticity of the user is comprehensively evaluated. This helps to more fully understand the behavior patterns of the user, improving accurate judgment of the user's authenticity. The functional relation calculates the authenticity score of the user through the multidimensional parameters, improves the accuracy and the comprehensiveness of the authenticity judgment, reduces the misjudgment rate and realizes comprehensive and personalized authenticity assessment. And the function relation is adjusted and corrected by using the abnormal adjustment value tau for optimizing the user authenticity score, so that the error influence caused by abnormal data or error items is reduced, the optimized user authenticity score P is generated more accurately, and the accuracy and reliability of optimizing the user authenticity judging mechanism by the user authenticity judging mechanism are improved. Meanwhile, the weight information and the adjustment value in the formula can be adjusted according to actual conditions and are applied to different parameters of a user authenticity judging mechanism, so that the flexibility and applicability of the algorithm are improved.
Preferably, step S4 comprises the steps of:
step S41: homomorphic encryption processing is carried out on the secure electronic commerce data by utilizing homomorphic encryption technology, and encrypted secure electronic commerce data is generated;
step S42: carrying out real-time updating processing on the secure electronic commerce data to generate real-time secure electronic commerce data;
step S43: and carrying out encryption data real-time updating on the encryption security electronic commerce data according to the real-time security electronic commerce protection data to generate real-time encryption security electronic commerce data.
The invention encrypts the secure electronic commerce data by using the homomorphic encryption technology to generate the encrypted secure electronic commerce data, wherein homomorphic encryption is a special encryption technology, which allows calculation operation in an encryption state without decrypting the data, and the privacy and confidentiality of the secure electronic commerce data are protected by homomorphic encryption, so that sensitive information is ensured not to be exposed in the processing process. The encryption security electronic commerce data is updated in real time to generate the real-time security electronic commerce data, the electronic commerce data may be changed continuously, such as updating of transaction data, modification of user information and the like, and through the real-time updating, the system can update the encryption security electronic commerce data in time, keep the latest state of the data, ensure that the updated data is still in an encryption state and sensitive information cannot be leaked. The encryption security electronic commerce data is updated in real time according to the real-time security electronic commerce protection data, the real-time encryption security electronic commerce data is generated, the real-time security electronic commerce protection data possibly contains the latest protection strategy, rules and feedback information of abnormal conditions, and the system can dynamically adjust the protection strategy and the processing mode of the data by applying the real-time security electronic commerce protection data to the encryption security electronic commerce data so as to timely cope with the newly-appearing security threats and abnormal conditions.
In this specification, there is provided an artificial intelligence electronic commerce data protection system for performing the artificial intelligence electronic commerce data protection method as described above, the artificial intelligence electronic commerce data protection system comprising:
the electronic commerce data processing module is used for acquiring original electronic commerce data; carrying out data preprocessing on the original electronic commerce data to generate standard electronic commerce data;
the electronic commerce data anomaly judgment module is used for carrying out electronic commerce data anomaly detection processing on the standard electronic commerce data to generate electronic commerce data anomaly values; data division is carried out on the standard electronic commerce data according to the electronic commerce data abnormal value, and abnormal electronic commerce data and conventional electronic commerce data are respectively generated;
the electronic commerce data protection module is used for carrying out abnormal field extraction processing on abnormal electronic commerce data to generate abnormal field data; e-business protection decision design is carried out according to the abnormal field data, and E-business protection decisions are generated; e-commerce data protection processing is carried out on conventional E-commerce data by utilizing an E-commerce protection decision, and safe E-commerce data is generated;
and the electronic commerce data encryption module is used for carrying out homomorphic encryption processing on the secure electronic commerce data by utilizing homomorphic encryption technology, and carrying out real-time updating to generate the real-time encrypted secure electronic commerce data.
The application has the beneficial effects that the application performs effective time sequence screening, target data screening, data cleaning and standardization processing on the original electronic commerce data, is beneficial to removing invalid data and unifying data formats, and provides high-quality data basis for subsequent anomaly detection and protection decision. The standard electronic commerce data is processed by using an anomaly detection algorithm, an anomaly value is identified, and whether the data belongs to anomalies is judged according to a set anomaly threshold value, so that the accuracy and the instantaneity of anomaly detection can be improved, and potential anomalies can be found early. The method is beneficial to deep analysis of the nature and the source of the abnormal behavior and provides detailed data support for subsequent protection decisions by carrying out feature extraction and attack type identification on the abnormal electronic commerce data. The user authenticity judging mechanism is designed, and protection schemes aiming at different attack types are formulated, so that the reliability of user identity verification and the targeted protection strategy can be improved, the safety of e-commerce data is comprehensively ensured, the targeted protection strategy can be formulated, the error processing of normal data is avoided, and the accuracy and the effectiveness of protection decision are improved. By means of anomaly detection, attack identification and protection scheme design, different types of anomalies and attack behaviors can be identified and handled quickly, safety and reliability of e-commerce data are improved, and users and e-commerce platforms are protected from potential risks. The method has the advantages that the homomorphic encryption technology is utilized to encrypt the safe electronic commerce data, the data is updated in real time, the latest state of the data is kept, the privacy and confidentiality of the data are protected, the instantaneity and the accuracy of the data are guaranteed, the accuracy and the instantaneity of the data are kept, the quick response to new security threats and abnormal conditions are facilitated, the timeliness of data protection is enhanced, the safety and the integrity of the electronic commerce data are comprehensively protected, the trust and the reputation of an electronic commerce platform are improved, and the sustainable development of business is promoted.
Drawings
FIG. 1 is a schematic flow chart of steps of an e-commerce data protection method of artificial intelligence according to the present invention;
FIG. 2 is a flowchart illustrating the detailed implementation of step S3 in FIG. 1;
FIG. 3 is a flowchart illustrating the detailed implementation of step S31 in FIG. 2;
FIG. 4 is a flowchart illustrating the detailed implementation of step S36 in FIG. 2;
the achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
The following is a clear and complete description of the technical method of the present patent in conjunction with the accompanying drawings, and it is evident that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, are intended to fall within the scope of the present invention.
Furthermore, the drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. The functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor methods and/or microcontroller methods.
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.
To achieve the above objective, referring to fig. 1 to 4, the present invention provides an artificial intelligence e-commerce data protection method, which includes the following steps:
step S1: acquiring original E-commerce data; carrying out data preprocessing on the original electronic commerce data to generate standard electronic commerce data;
step S2: performing electronic commerce data anomaly detection processing on standard electronic commerce data to generate electronic commerce data anomaly values; data division is carried out on the standard electronic commerce data according to the electronic commerce data abnormal value, and abnormal electronic commerce data and conventional electronic commerce data are respectively generated;
step S3: performing abnormal field extraction processing on the abnormal electronic commerce data to generate abnormal field data; e-business protection decision design is carried out according to the abnormal field data, and E-business protection decisions are generated; e-commerce data protection processing is carried out on conventional E-commerce data by utilizing an E-commerce protection decision, and safe E-commerce data is generated;
Step S4: and carrying out homomorphic encryption processing on the secure electronic commerce data by utilizing homomorphic encryption technology, and carrying out real-time updating to generate real-time encrypted secure electronic commerce data.
In the E-commerce field, the original data are usually from a plurality of different data sources, and the problems of non-uniform data format, deletion, errors and the like possibly exist, and the original E-commerce data can be cleaned, the deletion value is filled and the errors are corrected through data preprocessing, so that the standard E-commerce data are generated, the data quality is improved, the errors in subsequent processing are reduced, and the effectiveness and the accuracy of the subsequent steps are ensured. The data abnormality of the standard electronic commerce data is possibly caused by unexpected errors, malicious attacks or system faults, potential abnormal data can be found in time through electronic commerce data abnormality detection processing, abnormal electronic commerce data and conventional electronic commerce data are further separated, special processing and analysis are conducted on the abnormal data, interference of the abnormal data on the conventional data is reduced, and data processing efficiency is improved. The extraction of the abnormal field data can help reveal specific reasons for causing the abnormality, so that potential data security threat can be better known, and a targeted protection strategy and rule can be formulated according to the abnormal field data through the design of an e-commerce protection decision so as to cope with various abnormal conditions, improve the security of an e-commerce system and prevent potential risks of data leakage, tampering, theft and the like. The homomorphic encryption technology is an advanced encryption method, which allows data operation to be performed in an encryption state, and results are still kept encrypted, so that the technology enables the data to keep the encryption state in the transmission and processing processes, the confidentiality of the data can be protected even if the data is attacked, the security of the data is ensured to be continuously maintained when the data is updated in real time, security measures can be timely applied even if the data changes, and the homomorphic encryption can effectively protect the security of e-commerce data in the transmission and processing processes by combining the operation of the real-time updating. Therefore, the artificial intelligence electronic commerce data protection method can be used for analyzing historical attack means to relate to protection decisions when the electronic commerce data protection method is used for complex attack means, so that abnormal behaviors are analyzed and attack trends are predicted, and hidden attack means can be identified, so that the transaction safety of electronic commerce is ensured.
In the embodiment of the present invention, as described with reference to fig. 1, the step flow diagram of an electronic commerce data protection method of artificial intelligence of the present invention is provided, and in the embodiment, the electronic commerce data protection method of artificial intelligence includes the following steps:
step S1: acquiring original E-commerce data; carrying out data preprocessing on the original electronic commerce data to generate standard electronic commerce data;
in the embodiment of the invention, the original transaction data including user information, commodity information, transaction time and the like is acquired from the electronic commerce platform. And performing data cleaning, duplication removal and missing value processing on the original data, and regularizing the data into a standard format to generate standard E-commerce data.
Step S2: performing electronic commerce data anomaly detection processing on standard electronic commerce data to generate electronic commerce data anomaly values; data division is carried out on the standard electronic commerce data according to the electronic commerce data abnormal value, and abnormal electronic commerce data and conventional electronic commerce data are respectively generated;
in the embodiment of the invention, the standard electronic commerce data is subjected to electronic commerce data anomaly detection processing, such as detection algorithm detection is utilized for detecting commodity frequently traded by the same user in the standard electronic commerce data, users with abnormal transaction accounts and the like, and the abnormal value scoring of the data in the standard electronic commerce data is judged, so that the abnormal value of the electronic commerce data is generated. The standard electronic commerce data is divided into abnormal electronic commerce data and regular electronic commerce data, for example, standard electronic commerce data with abnormal values having scores of more than 80 points are marked as abnormal electronic commerce data, and standard electronic commerce data with abnormal values having scores of not more than 80 points are marked as regular electronic commerce data.
Step S3: performing abnormal field extraction processing on the abnormal electronic commerce data to generate abnormal field data; e-business protection decision design is carried out according to the abnormal field data, and E-business protection decisions are generated; e-commerce data protection processing is carried out on conventional E-commerce data by utilizing an E-commerce protection decision, and safe E-commerce data is generated;
in the embodiment of the invention, the abnormal electronic commerce data is subjected to abnormal field extraction processing, for example, fields with special abnormal characteristics such as abnormal transaction amount, abnormal login location and the like can be extracted from the abnormal data through a data mining technology. According to the extracted abnormal field data, we make an e-commerce protection decision, when the specified abnormal transaction amount exceeds a certain value, secondary verification, verification and evaluation of user identity are required, e-commerce data protection processing is carried out on conventional e-commerce data by using the e-commerce protection decision, for example, each e-commerce data is protected, and when a user accesses the e-commerce data or the transaction, the e-commerce protection decision is carried out, so that safe e-commerce data is generated.
Step S4: and carrying out homomorphic encryption processing on the secure electronic commerce data by utilizing homomorphic encryption technology, and carrying out real-time updating to generate real-time encrypted secure electronic commerce data.
In the embodiment of the invention, the homomorphic encryption technology is adopted to encrypt the secure electronic commerce data, the Paillier homomorphic encryption algorithm can be used to encrypt the transaction amount, the privacy information of the user is protected, and a real-time update mechanism is required to be established, for example, the encrypted secure electronic commerce data is updated at intervals, so that the real-time property and accuracy of the data are ensured.
Preferably, step S1 comprises the steps of:
step S11: acquiring original E-commerce data;
step S12: e-commerce data effective time sequence screening is carried out on original E-commerce data, and effective E-commerce data is generated;
step S13: data screening is carried out on the effective electronic commerce data according to a preset protection data target, and target electronic commerce data is generated;
step S14: performing data cleaning processing on the target electronic commerce data to generate cleaning electronic commerce data;
step S15: and carrying out data standardization processing on the cleaning electronic commerce data by using a minimum-maximum standardization method to generate standard electronic commerce data.
The original electronic commerce data may come from different data sources, such as transaction records, user behaviors, commodity information and the like, and a data basis is provided for the subsequent method. In an e-commerce environment, the effective time range of data is usually limited, and by carrying out effective time sequence screening on original e-commerce data, outdated, invalid or no longer needed data can be eliminated, so that the scale and complexity of a data set are reduced, the efficiency of subsequent processing is improved, the storage space is saved, and the timeliness of the data is ensured. The data subjected to effective time sequence screening is further screened according to a preset protection data target, such as protecting user privacy, preventing fraud, monitoring abnormal transactions and the like, focusing on important parts needing special protection in the e-commerce data, reducing the processing of irrelevant data and optimizing the data processing flow. The original data may contain problems such as noise, errors, missing values and the like, and the problems may cause errors and inaccuracy in the subsequent data processing process, and the errors in the data are corrected through the data cleaning process, so that the missing values are filled, the noise is eliminated, the quality and consistency of the data are ensured, and the reliability and accuracy of the subsequent processing steps are improved. The scales and ranges of different data may be different, which makes it difficult to effectively compare and analyze the data, and the cleaned e-commerce data is subjected to standardized processing by a min-max standardization method, so that the data is in a uniform scale range, and subsequent analysis and processing are facilitated, which helps to ensure comparability among different data features, and meanwhile, improves the efficiency of data processing and analysis.
In the embodiment of the invention, the original transaction data is obtained from the database of the e-commerce platform. The data includes purchase records of the user, sales information of the commodity, transaction time and the like, and for example, the transaction records of the commodity X purchased by the user A are extracted from a server of the electronic commerce platform, and the transaction records include purchase time, purchase quantity, commodity price and the like. And (3) carrying out time sequence screening on the original electronic commerce data, only keeping the data in an effective time range, for example, setting a time window, only keeping the transaction data in the last year, and eliminating the historical data after time. And screening the effective electronic commerce data according to a preset protection data target, and reserving data related to the target, for example, the protection target is used for detecting transaction behaviors, so that only transaction data related to high-value commodities are reserved, and transaction data of low-value commodities are filtered. And (3) performing data cleaning processing on the target electronic commerce data to remove incomplete, repeated or erroneous data, for example, checking whether the data has a missing value or an abnormal value, and supplementing or deleting the missing data to ensure the integrity and accuracy of the data. The data of the cleaning electronic commerce is standardized by using a minimum-maximum standardization method, namely the data is scaled according to a certain proportion, so that the value of the data is between 0 and 1, for example, the data with different dimensions such as commodity price and purchase quantity are standardized, and the data with different attributes are compared on the same scale, so that the subsequent data processing and analysis are convenient.
Preferably, step S2 comprises the steps of:
step S21: carrying out electronic commerce data anomaly detection processing on standard electronic commerce data by utilizing an electronic commerce data anomaly detection algorithm to generate electronic commerce data anomaly values;
step S22: e-commerce data abnormal value is judged according to a preset E-commerce data abnormal threshold value, and when the E-commerce data abnormal value is larger than the E-commerce data abnormal threshold value, standard E-commerce data corresponding to the E-commerce data abnormal value is marked as abnormal E-commerce data; and when the abnormal value of the electronic commerce data is not larger than the abnormal threshold value of the electronic commerce data, marking the standard electronic commerce data corresponding to the abnormal value of the electronic commerce data as conventional electronic commerce data.
The invention processes the standard electronic commerce data by utilizing the electronic commerce data anomaly detection algorithm to identify the possible anomaly value, and the system can discover the possible fraudulent transaction, the possible anomaly behavior or other potential risk conditions by carrying out anomaly detection on the standard electronic commerce data, which is helpful for discovering and identifying the anomaly condition in an early stage and provides important basis for subsequent data processing and protection decision. Judging an abnormal value of the electronic commerce data according to a preset electronic commerce data abnormal threshold value, and marking the corresponding standard electronic commerce data as abnormal electronic commerce data if the abnormal value of the electronic commerce data is larger than the electronic commerce data abnormal threshold value; in contrast, if the e-commerce data outlier is not greater than the e-commerce data outlier threshold, the corresponding standard e-commerce data is marked as regular e-commerce data. Through the judging and marking process, the system can effectively distinguish abnormal and normal E-commerce data, thereby providing important guidance for subsequent protection decisions.
In the embodiment of the invention, the standard electronic commerce data is processed by using an electronic commerce data anomaly detection algorithm, the electronic commerce data anomaly detection algorithm calculates an electronic commerce data anomaly value through a multidimensional parameter, such as an anomaly behavior existing in a transaction, or considers whether a user account is anomaly or not, and the like, and the anomaly score of each data point can be calculated through the algorithm, wherein the anomaly score reflects the difference degree of the data point and other data points. Performing abnormality judgment according to a preset e-commerce data abnormality threshold, for example, setting the score of the abnormality threshold to 80 points, and if the abnormality score of a certain data point is greater than 80 points, marking the data point as abnormal e-commerce data; otherwise, if the abnormal score is not more than 80, the data point is marked as conventional electronic commerce data, through the division, the abnormal electronic commerce data can be further processed and protected in a targeted manner, corresponding measures are taken to cope with potential safety risks and data threats, and meanwhile, the conventional electronic commerce data is kept in the original state, so that the data processing efficiency and the resource utilization rate are improved.
Preferably, the e-commerce data anomaly detection algorithm in step S21 is as follows:
Wherein K is expressed as an electronic commerce data abnormal value, N is expressed as a data category of standard electronic commerce data, y i An e-commerce data mean denoted as class i,the historical average value of the e-commerce data expressed as the i-th class is expressed as the comment similarity of the standard e-commerce data, V is expressed as the repetition degree of purchasing users of the standard e-commerce data, u is expressed as the transaction abnormal weight information of the standard e-commerce data, V is expressed as the order quantity of the standard e-commerce data, t is expressed as the time length of accessing commodities involved in accessing the standard e-commerce data, and delta is expressed as the abnormal adjustment value of the abnormal value of the e-commerce data.
The invention utilizes an electronic commerce data anomaly detection algorithm which comprehensively considers the data class N and the electronic commerce data mean value y of the ith class of standard electronic commerce data i Class i e-commerce data history meansThe method comprises the steps of commentary similarity o of standard electronic commerce data, repetition V of purchasing users of the standard electronic commerce data, abnormal transaction weight information u of the standard electronic commerce data, order quantity V of the standard electronic commerce data, commodity accessing time length t related to accessing the standard electronic commerce data and interaction relation among functions to form a functional relation:
that is to say,the data category of the standard e-commerce data represents the division of the e-commerce data on different categories, such as different commodity categories or different user behavior types; the average value of the e-commerce data of the i-th class is expressed as the average value of the standard e-commerce data on the class; the historical average of the e-commerce data of the i-th class represents the average of the data class in the historical record and is used for matching with the current standard e-commerce data Comparing the average values; the standard electronic commerce data comment similarity reflects the similarity degree of comment information in the standard electronic commerce data and can be used for evaluating the consensus degree of users on goods or services; the purchasing user repetition of the standard electronic commerce data is the purchasing repetition frequency of the user purchasing the same commodity in the standard electronic commerce data; the transaction anomaly weight information of the standard merchant data is used for giving weights with different anomaly degrees to the merchant data, and is related to different types of transaction anomalies; the order quantity of the standard e-commerce data reflects the transaction activity of the data class; the length of time for accessing the goods involved in accessing the standard electronic commerce data represents the time range of goods or services involved in accessing the standard electronic commerce data by the user, possibly reflecting the frequency of consumption behavior of the user. The function relation calculates abnormal values of the electronic commerce data through the multi-dimensional parameters, achieves comprehensive and personalized abnormal detection, provides a more accurate and effective analysis tool for electronic commerce data protection, helps an electronic commerce platform to find potential abnormal behaviors in time, and ensures data safety and business robust development. And the function relation is adjusted and corrected by utilizing the abnormal adjustment value delta of the abnormal value of the electronic commerce data, so that the error influence caused by abnormal data or error items is reduced, the abnormal value K of the electronic commerce data is more accurately generated, and the accuracy and the reliability of the electronic commerce data abnormal detection processing of the standard electronic commerce data are improved. Meanwhile, the weight information and the adjustment value in the formula can be adjusted according to actual conditions and are applied to different standard E-commerce data, so that the flexibility and applicability of the algorithm are improved.
Preferably, step S3 comprises the steps of:
step S31: carrying out characteristic extraction on the abnormal electronic commerce data to generate abnormal electronic commerce characteristic data;
step S32: carrying out log file tracking of the abnormal E-commerce characteristic data on the abnormal E-commerce characteristic data to generate an abnormal log file;
step S33: carrying out extraction processing on the abnormal field of the log file according to the abnormal log file to generate abnormal field data;
step S34: carrying out attack type identification of E-commerce data on the abnormal field data to generate attack type data;
step S35: performing exception access user field extraction on the exception field data to generate an exception access user field;
step S36: e-business protection decision design is carried out according to the attack type data and the abnormal access user field, and E-business protection decisions are generated;
step S37: and carrying out electronic commerce data protection processing on the conventional electronic commerce data by utilizing the electronic commerce protection decision to generate safe electronic commerce data.
The method and the device perform feature extraction on the data marked as the abnormal electronic commerce data, extract meaningful feature information such as transaction amount, transaction time, geographical position and the like from the abnormal electronic commerce data through the feature extraction, and the feature information is helpful for subsequent log file tracking and abnormal field extraction, so that a foundation for performing deep analysis on the abnormal data is provided. The abnormal E-commerce characteristic data is tracked by the log file, and the change and trend of the abnormal data are recorded in real time by writing the abnormal E-commerce characteristic data into the log file, so that the abnormal situation can be monitored in real time, and new abnormal modes and behaviors can be found in time. According to the method, the system can find the abnormal field, namely key information related to abnormal behaviors, such as abnormal transaction types, abnormal user IDs, abnormal IP addresses and the like, by analyzing the content in the log file, so that the system is helpful for further understanding abnormal conditions and provides detailed data support for subsequent attack type identification and E-commerce protection decision. The abnormal field data is analyzed to identify the attack type of the e-commerce data, the attack type identification can be performed based on technologies such as machine learning, deep learning and the like, different types of attack behaviors such as fraud, network attack and the like are identified through a training model, the source and the property of the abnormal data can be quickly and accurately determined, and a basis is provided for making a targeted e-commerce protection decision. The method comprises the steps of extracting an abnormal access user field from abnormal field data, and identifying an abnormal access user, namely a user suspected of participating in abnormal behaviors by analyzing user information in the abnormal field data, so that the method is beneficial to further tracking and monitoring sources of the abnormal behaviors and helping an e-commerce platform to take necessary protective measures. E-commerce protection decision design is carried out according to attack type data and abnormal access user fields, corresponding protection strategies are formulated according to the identified attack types and abnormal access users, for example, a risk scoring mechanism or transaction limits can be set for fraudulent transaction behaviors, and blocking treatment can be carried out for malicious attackers, so that targeted protection measures can be adopted for different abnormal conditions, and the safety and reliability of E-commerce data are improved. And the regular electronic commerce data is protected by utilizing an electronic commerce protection decision to generate safe electronic commerce data, and the regular electronic commerce data is safely processed according to a formulated protection strategy, so that the data is not influenced by potential abnormal behaviors, and the normal operation of an electronic commerce platform and the safety of user data are ensured.
As an example of the present invention, referring to fig. 2, a detailed implementation step flow diagram of step S3 in fig. 1 is shown, where step S3 includes:
step S31: carrying out characteristic extraction on the abnormal electronic commerce data to generate abnormal electronic commerce characteristic data;
in the embodiment of the invention, the abnormal electronic commerce data is subjected to feature extraction, for example, for the abnormal transaction data, average features such as transaction amount, commodity category, transaction time and the like can be extracted, and the abnormal electronic commerce data is converted into the abnormal electronic commerce feature data through feature extraction so as to facilitate subsequent processing and analysis.
Step S32: carrying out log file tracking of the abnormal E-commerce characteristic data on the abnormal E-commerce characteristic data to generate an abnormal log file;
in the embodiment of the invention, the abnormal electronic commerce characteristic data is tracked to generate the abnormal log file, for example, specific information and characteristics of the transaction, including transaction amount, commodity purchasing information and the like, are recorded for each abnormal transaction, and the abnormal log file is helpful for comprehensively knowing abnormal conditions and provides a powerful basis for subsequent protection decisions.
Step S33: carrying out extraction processing on the abnormal field of the log file according to the abnormal log file to generate abnormal field data;
in the embodiment of the invention, the abnormal field extraction processing is performed according to the abnormal log file to obtain the abnormal field data, for example, for the abnormal transaction log, the abnormal transaction amount, the abnormal commodity category and other field information are extracted from the abnormal field data, and the abnormal field data can help us to more accurately locate the abnormal condition and the abnormal characteristic.
Step S34: carrying out attack type identification of E-commerce data on the abnormal field data to generate attack type data;
in the embodiment of the invention, the attack type of the e-commerce data is identified for the abnormal field data, and the attack type data is generated, for example, a machine learning algorithm is used for identifying possible fraud types in the abnormal transaction data, such as false transaction, credit card theft and the like. The nature of the abnormal situation can be better understood through the identification of the attack type data, and guidance is provided for the subsequent protection decision.
Step S35: performing exception access user field extraction on the exception field data to generate an exception access user field;
in the embodiment of the invention, the abnormal access user field extraction is performed on the abnormal field data to obtain the abnormal access user field, for example, for an abnormal login log, the user ID and the equipment information of the abnormal login are extracted from the abnormal access user field, and the abnormal access user field helps us track the source of the abnormal behavior and the related user information.
Step S36: e-business protection decision design is carried out according to the attack type data and the abnormal access user field, and E-business protection decisions are generated;
in the embodiment of the invention, the design of the E-commerce protection decision is carried out according to the attack type data and the abnormal access user field. For example, for an attack of an abnormal transaction type, we can set a threshold to limit high-volume transactions; for the attack of abnormal login type, multi-factor authentication and other measures can be adopted.
Step S37: and carrying out electronic commerce data protection processing on the conventional electronic commerce data by utilizing the electronic commerce protection decision to generate safe electronic commerce data.
In the embodiment of the invention, abnormal conditions can be processed and protected in a targeted manner through the E-commerce protection decision, different E-commerce protection decisions are implemented for different E-commerce data, and the safe E-commerce data is generated.
Preferably, step S31 comprises the steps of:
step S311: carrying out e-commerce class data average value calculation on the conventional e-commerce data by using average value calculation, and generating classified conventional e-commerce average value data;
step S312: e-commerce data safety interval design is carried out according to classified conventional E-commerce mean value data, and a conventional E-commerce data interval is generated;
Step S313: and carrying out abnormal electronic commerce data extraction on the abnormal electronic commerce data based on the conventional electronic commerce data interval to generate abnormal electronic commerce characteristic data.
The invention processes the conventional electronic commerce data by using a mean value calculation method to generate a mean value of electronic commerce type data, wherein the electronic commerce type data refer to type data with similar characteristics, such as sales of different commodity types, the number of registered sources of users and the like, and the system can obtain typical performance of the type data by calculating the mean value of each electronic commerce type data and serve as a reference standard for subsequent anomaly detection. And designing a safety interval of the e-commerce data according to the classified conventional e-commerce mean value data, wherein the safety interval of the e-commerce data is a range, which represents the allowable variation range of the classified data, and the normal fluctuation range of the classified data is defined by setting the safety interval, so that the data exceeding the range may belong to abnormal conditions. And extracting abnormal electronic commerce data based on a conventional electronic commerce data interval, identifying data exceeding the electronic commerce data safety interval as abnormal electronic commerce data, and generating corresponding abnormal characteristic data which contain information exceeding a normal range, so that the subsequent abnormal detection and protection decision are facilitated.
As an example of the present invention, referring to fig. 3, a detailed implementation step flow diagram of step S31 in fig. 2 is shown, where step S31 includes:
step S311: carrying out e-commerce class data average value calculation on the conventional e-commerce data by using average value calculation, and generating classified conventional e-commerce average value data;
in the embodiment of the invention, the average value calculation is performed on the conventional electronic commerce data by using the average value calculation, for example, the indexes such as the average price, the average sales volume and the like of different types of commodities can be calculated according to the commodity types, and by the calculation, the classified conventional electronic commerce average value data can be obtained, and the data reflects the typical transaction conditions of the different types of commodities.
Step S312: e-commerce data safety interval design is carried out according to classified conventional E-commerce mean value data, and a conventional E-commerce data interval is generated;
in the embodiment of the invention, the design of the e-commerce data safety interval is performed according to the classified conventional e-commerce mean data, for example, the standard deviation and the confidence interval of each commodity category can be calculated to determine the reasonable range of commodity price and sales, and the conventional e-commerce data is limited in the reasonable range in order to establish a safety interval.
Step S313: and carrying out abnormal electronic commerce data extraction on the abnormal electronic commerce data based on the conventional electronic commerce data interval to generate abnormal electronic commerce characteristic data.
In the embodiment of the invention, based on the conventional electronic commerce data interval, the abnormal electronic commerce data is extracted, for example, the commodity price of a transaction is obviously beyond the range of the safety interval of the category, the commodity price is marked as the abnormal electronic commerce data, and potential abnormal transaction behaviors can be identified through the abnormal data extraction, so that an important basis is provided for the subsequent protection decision.
Preferably, step S36 comprises the steps of:
step S361: designing a user authenticity judging mechanism according to the abnormal access user field, and generating a user authenticity judging mechanism;
step S362: optimizing a user authenticity judging mechanism by using a user authenticity screening algorithm to generate an optimized user authenticity judging mechanism;
step S363: carrying out protection scheme design according to the attack type data to generate attack protection scheme data;
step S364: e-business protection decision integration is carried out according to the optimized user authenticity judgment mechanism and the protection scheme, and E-business protection decision is generated.
The invention designs a user authenticity judging mechanism according to the abnormal access user field, wherein the user authenticity judging mechanism aims at verifying the real identity of the user so as to identify the user with a potential malicious attacker or false identity, and the mechanism can comprise various verification means, such as short message verification codes, face recognition, voiceprint recognition and the like, so as to improve the reliability and the safety of user identity verification. The user authenticity judging mechanism is optimized by utilizing a user authenticity screening algorithm, the accuracy of judging the user authenticity is improved by analyzing historical data and behaviors, and the optimized user authenticity judging mechanism can better distinguish the real user from false identities and reduce fraudulent behaviors and attack behaviors. The design of the protection scheme is carried out according to attack type data, the attack type data provides detailed information about different attack behaviors, such as fraudulent transactions, malicious scripts and the like, and corresponding protection strategies, such as transaction risk scores, abnormal transaction detection, user behavior analysis and the like, are formulated according to the information so as to cope with the different attack behaviors. Integrating the optimized user authenticity judging mechanism with a protection scheme to generate an e-commerce protection decision, wherein the e-commerce protection decision comprehensively considers verification of user authenticity and different types of attack protection strategies to form a set of complete protection measures, and the decisions are applied to subsequent conventional e-commerce data protection processing to ensure the safety and reliability of e-commerce data.
As an example of the present invention, referring to fig. 4, a detailed implementation step flow diagram of step S36 in fig. 2 is shown, where step S36 includes:
step S361: designing a user authenticity judging mechanism according to the abnormal access user field, and generating a user authenticity judging mechanism;
in the embodiment of the invention, the design of the user authenticity judging mechanism is carried out according to the abnormal access user field, for example, a user behavior analysis model can be constructed, the authenticity of the user is judged by analyzing information such as the login place, login equipment, login frequency and the like of the user, and if the login place of a certain user suddenly changes across countries, the account of the user can be possibly stolen or abnormal behaviors exist.
Step S362: optimizing a user authenticity judging mechanism by using a user authenticity screening algorithm to generate an optimized user authenticity judging mechanism;
in the embodiment of the invention, the user authenticity judging mechanism is optimized by utilizing the user authenticity discriminating algorithm, for example, a machine learning-based algorithm can be adopted to analyze and predict the user behavior mode so as to improve the accuracy and precision of judging the user authenticity, and through optimization, the user can judge the user authenticity more accurately, and the situations of misjudgment and missed judgment are avoided.
Step S363: carrying out protection scheme design according to the attack type data to generate attack protection scheme data;
in the embodiment of the invention, the design of the protection scheme is carried out according to the attack type data, for example, for the discovered abnormal transaction type attack, the measures of risk assessment and transaction amount limitation are adopted; for the abnormal login type attack, protection measures such as multi-factor authentication and account verification are enhanced, and according to analysis of attack type data, a more targeted protection scheme can be formulated, so that the safety of electronic commerce data is improved.
Step S364: e-business protection decision integration is carried out according to the optimized user authenticity judgment mechanism and the protection scheme, and E-business protection decision is generated.
In the embodiment of the invention, the optimized user authenticity judging mechanism is integrated with the protection scheme to generate an e-commerce protection decision, for example, a threshold value of the user authenticity judging score can be set, when the user authenticity score is lower than the threshold value, corresponding protection measures are triggered, and meanwhile, the priority and the strictness degree of the protection scheme are adjusted according to attack type data. Through integration of the E-commerce protection decision, authenticity and potential attack types of users can be comprehensively considered, targeted protection measures are made, and safe operation of an E-commerce platform is ensured.
Preferably, the user authenticity screening algorithm in step S362 is as follows:
p represents the optimized user authenticity score, a represents the user e-commerce platform credit rating, r represents the network address anomaly weight information generated according to the network address accessing the e-commerce data, b represents the initial user authenticity score generated according to the user authenticity judging mechanism, m represents the proxy anomaly weight information generated according to the user agent accessing the e-commerce data, c represents the user e-commerce data access frequency, epsilon represents the user liveness score, d represents the user anomaly behavior score, tau represents the anomaly adjustment value optimizing the user authenticity score.
The invention utilizes a user authenticity screening algorithm which comprehensively considers the interaction relation among a user E-commerce platform credit rating a, network address abnormal weight information r generated according to a network address accessing E-commerce data, an initial user authenticity score b generated according to a user authenticity judging mechanism, agent abnormal weight information m generated according to a user agent accessing E-commerce data, a user E-commerce data access frequency c, a user liveness score epsilon, a user abnormal behavior score d and functions to form a functional relation:
That is to say,the credit rating of the user e-commerce platform represents the credit rating of the user on the e-commerce platform, which is an important index and reflects whether the user has bad records or credit problems in the past behaviors; the network address anomaly weight information generated according to the network address accessing the e-commerce data is expressed as being related to the trust degree of different network addresses used by the user; generation according to user authenticity judgment mechanismThe initial user authenticity score of (1) reflects the authenticity preliminary judgment result of the user; the agent abnormal weight information generated according to the user agent accessing the e-commerce data is used for reflecting the credibility degree of the user agent; the data access frequency of the user electronic commerce is expressed as the activity of the user and is related to the use habit and behavior of the user; the user liveness score is expressed as being obtained based on the access time length and the access frequency of the user on the e-commerce platform; the user abnormal behavior score is used to reflect whether the user has abnormal or suspicious behavior. The behavior characteristics (such as access frequency, abnormal behavior and the like) of the functional relation type user are comprehensively considered with information such as credit rating and the like, so that the authenticity of the user is comprehensively evaluated. This helps to more fully understand the behavior patterns of the user, improving accurate judgment of the user's authenticity. The functional relation calculates the authenticity score of the user through the multidimensional parameters, improves the accuracy and the comprehensiveness of the authenticity judgment, reduces the misjudgment rate and realizes comprehensive and personalized authenticity assessment. And the function relation is adjusted and corrected by using the abnormal adjustment value tau for optimizing the user authenticity score, so that the error influence caused by abnormal data or error items is reduced, the optimized user authenticity score P is generated more accurately, and the accuracy and reliability of optimizing the user authenticity judging mechanism by the user authenticity judging mechanism are improved. Meanwhile, the weight information and the adjustment value in the formula can be adjusted according to actual conditions and are applied to different parameters of a user authenticity judging mechanism, so that the flexibility and applicability of the algorithm are improved.
Preferably, step S4 comprises the steps of:
step S41: homomorphic encryption processing is carried out on the secure electronic commerce data by utilizing homomorphic encryption technology, and encrypted secure electronic commerce data is generated;
step S42: carrying out real-time updating processing on the secure electronic commerce data to generate real-time secure electronic commerce data;
step S43: and carrying out encryption data real-time updating on the encryption security electronic commerce data according to the real-time security electronic commerce protection data to generate real-time encryption security electronic commerce data.
The invention encrypts the secure electronic commerce data by using the homomorphic encryption technology to generate the encrypted secure electronic commerce data, wherein homomorphic encryption is a special encryption technology, which allows calculation operation in an encryption state without decrypting the data, and the privacy and confidentiality of the secure electronic commerce data are protected by homomorphic encryption, so that sensitive information is ensured not to be exposed in the processing process. The encryption security electronic commerce data is updated in real time to generate the real-time security electronic commerce data, the electronic commerce data may be changed continuously, such as updating of transaction data, modification of user information and the like, and through the real-time updating, the system can update the encryption security electronic commerce data in time, keep the latest state of the data, ensure that the updated data is still in an encryption state and sensitive information cannot be leaked. The encryption security electronic commerce data is updated in real time according to the real-time security electronic commerce protection data, the real-time encryption security electronic commerce data is generated, the real-time security electronic commerce protection data possibly contains the latest protection strategy, rules and feedback information of abnormal conditions, and the system can dynamically adjust the protection strategy and the processing mode of the data by applying the real-time security electronic commerce protection data to the encryption security electronic commerce data so as to timely cope with the newly-appearing security threats and abnormal conditions.
In the embodiment of the invention, homomorphic encryption technology is utilized to encrypt the secure electronic commerce data, for example, homomorphic encryption algorithm is used for encrypting personal information and transaction records of users, homomorphic encryption is a special encryption technology, data calculation and operation can be carried out in an encryption state without decrypting the data, and even if the data is operated in the encryption state, sensitive information of users can not be revealed. The method comprises the steps of carrying out real-time update processing on secure electronic commerce data, for example, the secure data can be continuously changed along with the transaction activities of a user on an electronic commerce platform, carrying out real-time monitoring and update on the secure data to keep the latest state of the data, for example, when the user carries out a transaction, the related transaction amount, purchased goods and other information can be changed, and the information needs to be updated in time to ensure real-time data protection and security, carrying out real-time update on the encrypted secure electronic commerce data according to the real-time secure electronic commerce protection data, for example, when the electronic commerce platform finds abnormal transactions or user login behaviors, triggering real-time secure electronic commerce protection measures, carrying out corresponding update and adjustment on the encrypted secure electronic commerce data according to the conditions of the protection data, and being capable of responding to potential security threats in time to ensure the security and integrity of the electronic commerce data.
In this specification, there is provided an artificial intelligence electronic commerce data protection system for performing the artificial intelligence electronic commerce data protection method as described above, the artificial intelligence electronic commerce data protection system comprising:
the electronic commerce data processing module is used for acquiring original electronic commerce data; carrying out data preprocessing on the original electronic commerce data to generate standard electronic commerce data;
the electronic commerce data anomaly judgment module is used for carrying out electronic commerce data anomaly detection processing on the standard electronic commerce data to generate electronic commerce data anomaly values; data division is carried out on the standard electronic commerce data according to the electronic commerce data abnormal value, and abnormal electronic commerce data and conventional electronic commerce data are respectively generated;
the electronic commerce data protection module is used for carrying out abnormal field extraction processing on abnormal electronic commerce data to generate abnormal field data; e-business protection decision design is carried out according to the abnormal field data, and E-business protection decisions are generated; e-commerce data protection processing is carried out on conventional E-commerce data by utilizing an E-commerce protection decision, and safe E-commerce data is generated;
and the electronic commerce data encryption module is used for carrying out homomorphic encryption processing on the secure electronic commerce data by utilizing homomorphic encryption technology, and carrying out real-time updating to generate the real-time encrypted secure electronic commerce data.
The application has the beneficial effects that the application performs effective time sequence screening, target data screening, data cleaning and standardization processing on the original electronic commerce data, is beneficial to removing invalid data and unifying data formats, and provides high-quality data basis for subsequent anomaly detection and protection decision. The standard electronic commerce data is processed by using an anomaly detection algorithm, an anomaly value is identified, and whether the data belongs to anomalies is judged according to a set anomaly threshold value, so that the accuracy and the instantaneity of anomaly detection can be improved, and potential anomalies can be found early. The method is beneficial to deep analysis of the nature and the source of the abnormal behavior and provides detailed data support for subsequent protection decisions by carrying out feature extraction and attack type identification on the abnormal electronic commerce data. The user authenticity judging mechanism is designed, and protection schemes aiming at different attack types are formulated, so that the reliability of user identity verification and the targeted protection strategy can be improved, the safety of e-commerce data is comprehensively ensured, the targeted protection strategy can be formulated, the error processing of normal data is avoided, and the accuracy and the effectiveness of protection decision are improved. By means of anomaly detection, attack identification and protection scheme design, different types of anomalies and attack behaviors can be identified and handled quickly, safety and reliability of e-commerce data are improved, and users and e-commerce platforms are protected from potential risks. The method has the advantages that the homomorphic encryption technology is utilized to encrypt the safe electronic commerce data, the data is updated in real time, the latest state of the data is kept, the privacy and confidentiality of the data are protected, the instantaneity and the accuracy of the data are guaranteed, the accuracy and the instantaneity of the data are kept, the quick response to new security threats and abnormal conditions are facilitated, the timeliness of data protection is enhanced, the safety and the integrity of the electronic commerce data are comprehensively protected, the trust and the reputation of an electronic commerce platform are improved, and the sustainable development of business is promoted.
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.
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 (10)

1. An artificial intelligence electronic commerce data protection method is characterized by comprising the following steps:
step S1: acquiring original E-commerce data; carrying out data preprocessing on the original electronic commerce data to generate standard electronic commerce data;
step S2: performing electronic commerce data anomaly detection processing on standard electronic commerce data to generate electronic commerce data anomaly values; data division is carried out on the standard electronic commerce data according to the electronic commerce data abnormal value, and abnormal electronic commerce data and conventional electronic commerce data are respectively generated;
Step S3: performing abnormal field extraction processing on the abnormal electronic commerce data to generate abnormal field data; e-business protection decision design is carried out according to the abnormal field data, and E-business protection decisions are generated; e-commerce data protection processing is carried out on conventional E-commerce data by utilizing an E-commerce protection decision, and safe E-commerce data is generated;
step S4: and carrying out homomorphic encryption processing on the secure electronic commerce data by utilizing homomorphic encryption technology, and carrying out real-time updating to generate real-time encrypted secure electronic commerce data.
2. The method for protecting e-commerce data of artificial intelligence according to claim 1, wherein the step S1 comprises the steps of:
step S11: acquiring original E-commerce data;
step S12: e-commerce data effective time sequence screening is carried out on original E-commerce data, and effective E-commerce data is generated;
step S13: data screening is carried out on the effective electronic commerce data according to a preset protection data target, and target electronic commerce data is generated;
step S14: performing data cleaning processing on the target electronic commerce data to generate cleaning electronic commerce data;
step S15: and carrying out data standardization processing on the cleaning electronic commerce data by using a minimum-maximum standardization method to generate standard electronic commerce data.
3. The method for protecting e-commerce data of artificial intelligence according to claim 2, wherein the step S2 comprises the steps of:
Step S21: carrying out electronic commerce data anomaly detection processing on standard electronic commerce data by utilizing an electronic commerce data anomaly detection algorithm to generate electronic commerce data anomaly values;
step S22: e-commerce data abnormal value is judged according to a preset E-commerce data abnormal threshold value, and when the E-commerce data abnormal value is larger than the E-commerce data abnormal threshold value, standard E-commerce data corresponding to the E-commerce data abnormal value is marked as abnormal E-commerce data; and when the abnormal value of the electronic commerce data is not larger than the abnormal threshold value of the electronic commerce data, marking the standard electronic commerce data corresponding to the abnormal value of the electronic commerce data as conventional electronic commerce data.
4. The method for protecting e-commerce data of artificial intelligence according to claim 3, wherein the algorithm for detecting e-commerce data anomalies in step S21 is as follows:
wherein K is expressed as an electronic commerce data abnormal value, N is expressed as a data category of standard electronic commerce data, y i An e-commerce data mean denoted as class i,the historical average value of the e-commerce data expressed as the i-th class is expressed as the comment similarity of the standard e-commerce data, V is expressed as the repetition degree of purchasing users of the standard e-commerce data, u is expressed as the transaction abnormal weight information of the standard e-commerce data, V is expressed as the order quantity of the standard e-commerce data, t is expressed as the time length of accessing commodities involved in accessing the standard e-commerce data, and delta is expressed as the abnormal adjustment value of the abnormal value of the e-commerce data.
5. The method for protecting e-commerce data of artificial intelligence according to claim 4, wherein the step S3 comprises the steps of:
step S31: carrying out characteristic extraction on the abnormal electronic commerce data to generate abnormal electronic commerce characteristic data;
step S32: carrying out log file tracking of the abnormal E-commerce characteristic data on the abnormal E-commerce characteristic data to generate an abnormal log file;
step S33: carrying out extraction processing on the abnormal field of the log file according to the abnormal log file to generate abnormal field data;
step S34: carrying out attack type identification of E-commerce data on the abnormal field data to generate attack type data;
step S35: performing exception access user field extraction on the exception field data to generate an exception access user field;
step S36: e-business protection decision design is carried out according to the attack type data and the abnormal access user field, and E-business protection decisions are generated;
step S37: and carrying out electronic commerce data protection processing on the conventional electronic commerce data by utilizing the electronic commerce protection decision to generate safe electronic commerce data.
6. The method for protecting e-commerce data of artificial intelligence according to claim 5, wherein the step S31 comprises the steps of:
Step S311: carrying out e-commerce class data average value calculation on the conventional e-commerce data by using average value calculation, and generating classified conventional e-commerce average value data;
step S312: e-commerce data safety interval design is carried out according to classified conventional E-commerce mean value data, and a conventional E-commerce data interval is generated;
step S313: and carrying out abnormal electronic commerce data extraction on the abnormal electronic commerce data based on the conventional electronic commerce data interval to generate abnormal electronic commerce characteristic data.
7. The method of claim 6, wherein step S36 includes the steps of:
step S361: designing a user authenticity judging mechanism according to the abnormal access user field, and generating a user authenticity judging mechanism;
step S362: optimizing a user authenticity judging mechanism by using a user authenticity screening algorithm to generate an optimized user authenticity judging mechanism;
step S363: carrying out protection scheme design according to the attack type data to generate attack protection scheme data;
step S364: e-business protection decision integration is carried out according to the optimized user authenticity judgment mechanism and the protection scheme, and E-business protection decision is generated.
8. The method for protecting e-commerce data of artificial intelligence of claim 7, wherein the user authenticity discrimination algorithm in step S362 is as follows:
P represents the optimized user authenticity score, a represents the user e-commerce platform credit rating, r represents the network address anomaly weight information generated according to the network address accessing the e-commerce data, b represents the initial user authenticity score generated according to the user authenticity judging mechanism, m represents the proxy anomaly weight information generated according to the user agent accessing the e-commerce data, c represents the user e-commerce data access frequency, epsilon represents the user liveness score, d represents the user anomaly behavior score, tau represents the anomaly adjustment value optimizing the user authenticity score.
9. The method of claim 8, wherein step S4 includes the steps of:
step S41: homomorphic encryption processing is carried out on the secure electronic commerce data by utilizing homomorphic encryption technology, and encrypted secure electronic commerce data is generated;
step S42: carrying out real-time updating processing on the secure electronic commerce data to generate real-time secure electronic commerce data;
step S43: and carrying out encryption data real-time updating on the encryption security electronic commerce data according to the real-time security electronic commerce protection data to generate real-time encryption security electronic commerce data.
10. An artificial intelligence e-commerce data protection system for performing the artificial intelligence e-commerce data protection method of claim 1, the artificial intelligence e-commerce data protection system comprising:
The electronic commerce data processing module is used for acquiring original electronic commerce data; carrying out data preprocessing on the original electronic commerce data to generate standard electronic commerce data;
the electronic commerce data anomaly judgment module is used for carrying out electronic commerce data anomaly detection processing on the standard electronic commerce data to generate electronic commerce data anomaly values; data division is carried out on the standard electronic commerce data according to the electronic commerce data abnormal value, and abnormal electronic commerce data and conventional electronic commerce data are respectively generated;
the electronic commerce data protection module is used for carrying out abnormal field extraction processing on abnormal electronic commerce data to generate abnormal field data; e-business protection decision design is carried out according to the abnormal field data, and E-business protection decisions are generated; e-commerce data protection processing is carried out on conventional E-commerce data by utilizing an E-commerce protection decision, and safe E-commerce data is generated;
and the electronic commerce data encryption module is used for carrying out homomorphic encryption processing on the secure electronic commerce data by utilizing homomorphic encryption technology, and carrying out real-time updating to generate the real-time encrypted secure electronic commerce data.
CN202311017169.3A 2023-08-12 2023-08-12 Artificial intelligence electronic commerce data protection method and system Withdrawn CN116976894A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311017169.3A CN116976894A (en) 2023-08-12 2023-08-12 Artificial intelligence electronic commerce data protection method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311017169.3A CN116976894A (en) 2023-08-12 2023-08-12 Artificial intelligence electronic commerce data protection method and system

Publications (1)

Publication Number Publication Date
CN116976894A true CN116976894A (en) 2023-10-31

Family

ID=88483127

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311017169.3A Withdrawn CN116976894A (en) 2023-08-12 2023-08-12 Artificial intelligence electronic commerce data protection method and system

Country Status (1)

Country Link
CN (1) CN116976894A (en)

Similar Documents

Publication Publication Date Title
Ahmed et al. False data injection attack (FDIA): an overview and new metrics for fair evaluation of its countermeasure
US12052218B2 (en) Systems and methods to secure API platforms
US8739278B2 (en) Techniques for fraud monitoring and detection using application fingerprinting
JP4954979B2 (en) Systems and methods for fraud monitoring, detection, and hierarchical user authentication
CN113542279B (en) Network security risk assessment method, system and device
US20200004989A1 (en) Systems and methods of determining compromised identity information
WO2019079708A1 (en) Enhanced system and method for identity evaluation using a global score value
CN112637108B (en) Internal threat analysis method and system based on anomaly detection and emotion analysis
CN118134634B (en) Internet credit integrated management system
CN117272349A (en) Method, system and storage medium for protecting security of relational database
Datta et al. Real-time threat detection in ueba using unsupervised learning algorithms
Kshetri et al. cryptoRAN: A review on cryptojacking and ransomware attacks wrt banking industry-threats, challenges, & problems
Sarabi et al. Prioritizing Security Spending: A Quantitative Analysis of Risk Distributions for Different Business Profiles.
CN116541815B (en) Computer equipment operation and maintenance data safety management system
CN111970272A (en) APT attack operation identification method
Liu A risk-based approach to cybersecurity: a case study of financial messaging networks data breaches
CN118368083A (en) Digital resource sharing and access control method
Mihailescu et al. Unveiling Threats: Leveraging User Behavior Analysis for Enhanced Cybersecurity
CN116976894A (en) Artificial intelligence electronic commerce data protection method and system
CN118138312B (en) Intelligent payment port encryption method and system
CN118536164B (en) Data blood-edge tracking method, system and device for privacy security protection
CN117473475B (en) Big data security protection method, system and medium based on trusted computing
Al-Ruwili et al. Analysis of Ransomware Impact on Android Systems using Machine Learning Techniques.
Jyothi et al. Challenges and Artificial Intelligence–Centered Defensive Strategies for Authentication in Online Banking
Alotibi et al. The feasibility of using behavioural profiling technique for mitigating insider threats

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WW01 Invention patent application withdrawn after publication
WW01 Invention patent application withdrawn after publication

Application publication date: 20231031