CN113988878B - Graph database technology-based anti-fraud method and system - Google Patents
Graph database technology-based anti-fraud method and system Download PDFInfo
- Publication number
- CN113988878B CN113988878B CN202111609100.0A CN202111609100A CN113988878B CN 113988878 B CN113988878 B CN 113988878B CN 202111609100 A CN202111609100 A CN 202111609100A CN 113988878 B CN113988878 B CN 113988878B
- Authority
- CN
- China
- Prior art keywords
- information
- data
- data set
- fraud
- user
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION 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
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
- G06Q20/401—Transaction verification
- G06Q20/4016—Transaction verification involving fraud or risk level assessment in transaction processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
- G06F16/9024—Graphs; Linked lists
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/906—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Linguistics (AREA)
- General Business, Economics & Management (AREA)
- Evolutionary Biology (AREA)
- Economics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Marketing (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Technology Law (AREA)
- Development Economics (AREA)
- Software Systems (AREA)
- Computer Security & Cryptography (AREA)
- Animal Behavior & Ethology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides an anti-fraud method and an anti-fraud system based on a graph database technology, wherein the method comprises the following steps: the method comprises the steps that based on data sharing of an anti-fraud analysis platform and a multi-party data source platform, multi-party data information of a user set is obtained, a user knowledge graph is constructed, and targeted retrieval is conducted on the user knowledge graph to obtain a retrieval data set; and carrying out normalization processing on the retrieval data set, carrying out clustering analysis, constructing a training data set according to a clustering result, carrying out abnormity detection on loan behaviors through an isolated forest method according to the training data set, obtaining abnormal loan information, obtaining anti-fraud reminding information, and reminding a user of the existence of abnormity of loan through the anti-fraud reminding information. The problems that in the prior art, due to the fact that data sources are numerous, data scale is large and the like, the anti-fraud technology is difficult to accurately verify data information, and fraud recognition capability is low are solved.
Description
Technical Field
The invention relates to the field of finance, in particular to an anti-fraud method and an anti-fraud system based on a graph database technology.
Background
The main risks of credit business are from operational risks, fraud risks and credit risks. Currently, loan fraud is becoming more "covert", and it is difficult for original big data anti-fraud wind control techniques to identify new types of fraud. The large data anti-fraud method is characterized in that the authenticity of information is cross-verified by collecting a large amount of heterogeneous and diversified information, but the problems of multiple data sources, heterogeneous fragmentation of data, coexistence of structural, semi-structural and unstructured data, increasingly huge data scale, integration and utilization of multiple heterogeneous data sources and the like exist, and along with increasingly diversified and concealed loan fraud modes, more and more banking institutions are aware of the limitation of the traditional large data anti-fraud technology. Graph databases are a type of NoSQL database, and may also be referred to as graph-oriented/based databases. The graph database technology can integrate multi-source heterogeneous big data, and converts the verification of the identity, the data and the like of points into a form of a surface to carry out fraud risk detection, thereby realizing the identification and the defense of fraud.
However, in the prior art, due to the problems of numerous data sources, huge data scale and the like, the anti-fraud technology is difficult to accurately verify data information, and the fraud identification capability is low.
Disclosure of Invention
The application provides an anti-fraud method and system based on a graph database technology, and solves the problems that in the prior art, due to the fact that data sources are numerous, data scale is large and the like, the anti-fraud technology is difficult to accurately verify data information, and fraud recognition capability is low. The method achieves the technical effects of improving the efficiency of risk analysis on user data through a graph database technology, and identifying risk business through an anomaly detection algorithm, so that the anti-fraud capacity of a bank is improved, and the identification accuracy of the loan of an abnormal user is improved.
In view of the above problems, the present application provides an anti-fraud method and system based on graph database technology.
In a first aspect, the present application provides a graph database technology-based anti-fraud method, including: obtaining multi-party data information of a user set through data sharing of the anti-fraud analysis platform and the multi-party data source platform; constructing a user knowledge graph according to the multi-party data information, and performing targeted retrieval on the user knowledge graph to obtain a retrieval data set; after normalization processing is carried out on the retrieval data set, clustering analysis is carried out, a training data set is constructed according to clustering results, abnormal loan behaviors are detected through an isolated forest method according to the training data set, abnormal loan information is obtained, first anti-fraud reminding information is obtained based on the abnormal loan information, and the user is reminded that the loan is abnormal through the first anti-fraud reminding information.
In another aspect, the present application provides a graph database technology-based anti-fraud system, the system comprising: the first obtaining unit is used for obtaining multi-party data information of a user set based on data sharing of the anti-fraud analysis platform and the multi-party data source platform; a first construction unit for constructing a user knowledge graph based on the multi-party data information; a second obtaining unit, configured to perform targeted retrieval on the user knowledge graph to obtain a retrieval data set; the second construction unit is used for carrying out normalization processing on the retrieval data set, then carrying out clustering analysis, and constructing a training data set according to a clustering result; the third obtaining unit is used for carrying out abnormal detection on loan behaviors through an isolated forest method according to the training data set to obtain abnormal loan information; and the fourth obtaining unit is used for obtaining first anti-fraud reminding information based on the abnormal loan information, and reminding the user that the loan is abnormal through the first anti-fraud reminding information.
In a third aspect, the present invention provides an anti-fraud system based on graph database technology, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the method of the first aspect when executing the program.
One or more technical solutions provided in the present application have at least the following technical effects or advantages:
due to the adoption of a multi-party data sharing mode, multi-party data information about a user set is obtained; constructing a user knowledge graph by using the multi-party data information; further retrieving the user knowledge graph to obtain a retrieval data set; the method comprises the steps of carrying out deep analysis on the retrieval data set, carrying out clustering analysis after normalization processing, constructing a training data set according to clustering results, carrying out abnormal detection on loan behaviors by using an isolated forest method, and obtaining first anti-fraud reminding information according to abnormal detection results, namely abnormal loan information, so as to carry out user loan abnormity reminding on a main body such as a bank.
The foregoing description is only an overview of the technical solutions of the present application, and the present application can be implemented according to the content of the description in order to make the technical means of the present application more clearly understood, and the following detailed description of the present application is given in order to make the above and other objects, features, and advantages of the present application more clearly understandable.
Drawings
FIG. 1 is a schematic flow chart illustrating an anti-fraud method based on graph database technology according to an embodiment of the present application;
FIG. 2 is a schematic diagram illustrating a process of constructing a training data set according to an anti-fraud method based on graph database technology according to an embodiment of the present application;
fig. 3 is a schematic flow chart illustrating the abnormal detection of loan behavior by an isolated forest method in an anti-fraud method based on a graph database technology according to an embodiment of the present application;
FIG. 4 is a schematic flow chart of a method for constructing a user knowledge graph based on graph database technology according to an embodiment of the present application;
FIG. 5 is a schematic diagram illustrating an anti-fraud system according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an exemplary electronic device according to an embodiment of the present application.
Description of reference numerals: the system comprises a first obtaining unit 11, a first constructing unit 12, a second obtaining unit 13, a second constructing unit 14, a third obtaining unit 15, a fourth obtaining unit 16, an electronic device 300, a memory 301, a processor 302, a communication interface 303 and a bus architecture 304.
Detailed Description
The application provides an anti-fraud method and system based on a graph database technology, and solves the problems that in the prior art, due to the fact that data sources are numerous, data scale is large and the like, the anti-fraud technology is difficult to accurately verify data information, and fraud recognition capability is low. The method achieves the technical effects of improving the efficiency of risk analysis on user data through a graph database technology, and identifying risk business through an anomaly detection algorithm, so that the anti-fraud capacity of a bank is improved, and the identification accuracy of the loan of an abnormal user is improved.
Currently, loan fraud is becoming more "covert," and it is difficult for original big data anti-fraud wind control techniques to identify new types of fraud. The large data anti-fraud method is characterized in that the authenticity of information is cross-verified by collecting a large amount of heterogeneous and diversified information, but the problems of multiple data sources, heterogeneous fragmentation of data, coexistence of structural, semi-structural and unstructured data, increasingly huge data scale, integration and utilization of multiple heterogeneous data sources and the like exist, and along with increasingly diversified and concealed loan fraud modes, more and more banking institutions are aware of the limitation of the traditional large data anti-fraud technology. In the prior art, the problems of numerous data sources, huge data scale and the like exist, so that the anti-fraud technology is difficult to accurately verify data information, and the identification capability of fraud is low.
In view of the above technical problems, the technical solution provided by the present application has the following general idea:
the application provides an anti-fraud method based on graph database technology, which comprises the following steps: obtaining multi-party data information of a user set through data sharing of the anti-fraud analysis platform and the multi-party data source platform; according to the multi-party data information, a user knowledge graph is constructed, and the user knowledge graph is searched to obtain a search data set; and after normalization processing is carried out on the retrieval data set, clustering analysis is carried out, a training data set is constructed according to a clustering result, abnormal loan behaviors are detected through an isolated forest method according to the training data set, abnormal loan information is obtained, first anti-fraud reminding information is obtained based on the abnormal loan information, and the user is reminded that the loan is abnormal through the first anti-fraud reminding information.
Having thus described the general principles of the present application, various non-limiting embodiments thereof will now be described in detail with reference to the accompanying drawings.
Example one
As shown in fig. 1, an embodiment of the present application provides an anti-fraud method based on graph database technology, where the method is applied to an anti-fraud analysis platform, the platform is communicatively connected to a multi-party data source platform, and the method includes:
s100: obtaining multi-party data information of a user set based on data sharing of the anti-fraud analysis platform and the multi-party data source platform;
further, the step S100 of obtaining multi-party data information of a user set based on the data sharing between the anti-fraud analysis platform and the multi-party data source platform includes:
s110: and acquiring multi-party data information of the user set based on the data sharing of the anti-fraud analysis platform and the multi-party data source platform, wherein the multi-party data information comprises user social relations, transaction mode association, internet behaviors and mobile equipment data characteristics.
Specifically, the anti-fraud analysis platform provides anti-fraud analysis services for a main body such as a bank. The multi-party data source platform is a public data platform capable of acquiring user data, such as China P2P network loan index, China financial information network, network loan data and the like. In order to obtain a large amount of user data and facilitate analysis of users, the anti-fraud analysis platform and the data of the multi-party data source platform are shared to obtain multi-party data information of a user set. Where the user includes a group, such as an individual, business, etc., that is conducting credit transactions with the bank. The multi-party data information of the user set comprises user social relations, transaction mode associations, internet behaviors and mobile device data characteristics. The user social relationship comprises relationship person information and relationship group information, such as relatives of the user, units of the user and the like. The transaction mode association is to associate different transaction modes, for example, transactions using a bank card, transactions using a mobile phone payment, transactions using cash, and the like. The internet behaviors comprise various behaviors performed by the user on the internet, such as performing multiple loans on loan software. The mobile device data characteristics comprise data information generated when a user uses the mobile device to conduct online transaction and online account transfer, and the conditions of credit loss people blacklisting people involved in social records and the like. And multi-party data information of the user set is obtained, so that risk analysis is convenient for the user, and a foundation is laid for anti-fraud of a bank system.
S200: constructing a user knowledge graph based on the multi-party data information;
further, the step S200 of constructing a user knowledge graph based on the multi-party data information includes:
s210: carrying out structured classification on the multi-party data information to obtain a structured data set, an unstructured data set and a semi-structured data set;
s220: and extracting knowledge from the unstructured data set and the semi-structured data set, and then performing knowledge fusion on the unstructured data set and the structured data set to construct the user knowledge graph.
Specifically, due to the difference in the structure of the multi-party data information, the multi-party data is structurally classified, and the multi-party data information is divided into a structured data set, an unstructured data set and a semi-structured data set. Structured data refers to data that can be represented and stored using a relational database, in two dimensions. For example: enterprise management information systems, financial systems, etc. The unstructured data set has an irregular or incomplete data structure and no predefined data model, so that the data represented by a database two-dimensional logic table is inconvenient. For example: pictures, various types of reports, images, audio/video information, and the like. Semi-structured data refers to data model structures that are not associated with a relational database or other data table format, but contain relevant markers to separate semantic elements and to stratify records and fields, such as: mail, HTML, report, resource library.
To construct the user knowledge graph using the structured dataset, it is necessary to extract knowledge from the unstructured dataset and the semi-structured dataset, in other words, extract entities and connections in the unstructured dataset and the semi-structured dataset, and obtain connections between the extracted entities, to name but not limited to: the unstructured data set and the semi-structured data set can be captured or analyzed through the existing computer algorithm to obtain entities and relations, and the data extraction process is completed, so that the extracted knowledge base is obtained.
The same entity may have different meanings in different scenes, ambiguity elimination is carried out through knowledge fusion, the entity and the relation are combined, redundancy and wrong concepts are removed, and the quality of knowledge in a knowledge base is guaranteed. And integrating entities with the entities in the structured data to complete a knowledge fusion process, and constructing the user knowledge graph based on the fused data set.
S300: performing targeted retrieval on the user knowledge graph to obtain a retrieval data set;
further, the step S300 of performing targeted retrieval on the user knowledge graph to obtain a retrieval data set includes:
s310: setting pre-calibrated information, wherein the pre-calibrated information comprises loan amount characteristics, money information, associated advertisement characteristics and character loophole characteristics, and direct or indirect relevance exists among the characteristics in the pre-calibrated information;
s320: and according to the pre-calibrated information, performing target retrieval through the user knowledge graph to obtain the retrieval data set.
Specifically, the preset target comprises loan amount characteristics, money information, associated advertisement characteristics and character loophole characteristics, wherein direct or indirect relevance exists among the characteristics in the preset target information, namely, the direct relevance characteristics and the indirect relevance characteristics related to the loan are extracted, and a retrieval data set is obtained through traversal retrieval. And the retrieval data set retrieves and extracts the characteristics of the bank loan part in the user knowledge map, thereby ensuring the comprehensiveness of information extraction and being beneficial to the bank to make loan risk decisions.
S400: carrying out normalization processing on the retrieval data set, then carrying out clustering analysis, and constructing a training data set according to a clustering result;
further, as shown in fig. 2, the step S400 includes performing a normalization process on the search data set, performing a cluster analysis, and constructing a training data set according to a clustering result, where:
s410: after normalization processing is carried out on the retrieval data set, a first retrieval data set is obtained;
s420: clustering and dividing the first retrieval data set to obtain a first clustering result, wherein the first clustering result comprises a plurality of data sets of different categories;
s430: and performing principal component analysis on the data sets of different categories to construct the training data set.
Specifically, the data of the retrieval data set is rich and complicated, and the data with different characteristics are normalized, so that the subsequent data can be analyzed and processed conveniently. To give an example without limitation: normalization from the perspective of aggregation, dimension normalization can be performed, that is, abstraction normalization is performed, attributes of elements in an unimportant and non-comparable aggregation are removed, and attributes related to topics are reserved, so that objects or things which do not have comparability originally can be normalized, and comparison can be performed after the objects or things are classified into one class. Meanwhile, different data can also be normalized through a representation method of relative quantity. From the perspective of data, normalization can also be viewed as a process of removing dimension, removing unit limits. After normalization processing, the first retrieval data set which is easy to analyze is obtained.
Further, clustering division is carried out on the first retrieval data set, and a first clustering result is obtained. The cluster analysis belongs to the category of unsupervised learning in machine learning, and can be but not limited to cluster division through a K-means clustering method, a clustering method according to density clustering, hierarchical clustering and the like to obtain the first clustering result, and divide the first retrieval data into a plurality of data sets of different categories. Performing data dimension reduction on different types of data sets through principal component analysis, taking a clustering analysis result as a feature data set in the process of principal component analysis, performing average value calculation, subtracting the average value of each sample to obtain a new feature value, performing calculation through a covariance formula to obtain a covariance matrix, obtaining the feature value and the feature vector of the covariance matrix, selecting the first K feature values with the largest maximum and the feature vectors corresponding to the first K feature values, projecting the original features in the feature data set onto the selected feature vectors, and obtaining the data set after dimension reduction, namely the training set. Redundant data volume can be eliminated under the condition of ensuring sufficient data volume through cluster analysis and principal component analysis, so that the sample volume of data operation is reduced, the minimum loss after dimension reduction is ensured, and the speed of data operation is improved.
S500: carrying out abnormal detection on loan behaviors through an isolated forest method according to the training data set to obtain abnormal loan information;
further, as shown in fig. 3, the step S500 of performing abnormal detection on loan behaviors by an isolated forest method according to the training data set to obtain abnormal loan information further includes:
s510: randomly non-back-placed drawn test sample sets from the training data set;
s520: randomly extracting first test data from the test sample set as a root node of an isolation tree;
s530: obtaining characteristic value information of the first test data;
s540: performing binary division on the test sample set according to the characteristic value information to obtain a left branch and a right branch, wherein the characteristic value of sample data in the left branch is smaller than the characteristic value information; the eigenvalue of the sample data in the right branch is greater than the eigenvalue information;
s550: repeating the binary division on the branches generated by the root node until a predetermined condition is reached, stopping the binary division, and taking the isolated branches as the abnormal loan information.
In particular, isolated forest algorithms are primarily directed to outliers in continuous structured data. The use of isolated forests presupposes that outliers are defined as outliers that are easily isolated. In order to construct an isolated tree, a feature and its segmentation value are randomly selected, and the data set is recursively segmented until any one of the following conditions is satisfied: the tree has reached a limited height; there is only one sample on a node; all features of the samples on the node are the same. In the training process of the isolated forest, each isolated tree randomly selects a part of samples, each tree is independently generated, and generally, the more the trees are, the more stable the algorithm is.
The method comprises the steps of extracting a test sample set in a random non-return mode, randomly extracting first test data in the test sample set, and putting the first test data into a root node of an isolation tree (isolation tree). Randomly appointing a dimension, obtaining characteristic value information of first test data which is randomly extracted as a cutting point, selecting the cutting point to generate a hyperplane, splitting a data space of a current node into two subspaces, namely performing binary division on the test sample set according to the characteristic value information, setting branches of a binary tree as a left branch and a right branch, placing a point which is less than the characteristic value in the left branch of the current node, and placing a point which is more than or equal to the characteristic value in the right branch of the current node. That is, the eigenvalue of the sample data in the left branch is smaller than the eigenvalue information; the eigenvalue of the sample data in the right branch is greater than the eigenvalue information.
The preset conditions are that the tree reaches the limited height and/or only one sample on the node and/or all the characteristics of the samples on the node are the same, when the preset conditions are reached, the binary division is stopped, and the isolated forest is obtained through the binary division of t isolated trees. And carrying out abnormal detection by using the generated isolated forest to obtain abnormal loan information. For each data point xiMake it traverse each isolated tree, calculate point xiAverage height h (x) in foresti) And normalizing the average height of all the points.
The formula for calculating the outlier score is as follows:
the abnormal loan information is obtained by carrying out abnormal detection on loan behaviors through an isolated forest method and identifying abnormality through an outlier in isolation data, so that a bank can be helped to accurately identify abnormal credit business.
S600: and acquiring first anti-fraud reminding information based on the abnormal loan information, and reminding the user of abnormal loan through the first anti-fraud reminding information.
Specifically, the abnormal loan information is formed into the first anti-fraud reminding information, the first anti-fraud reminding information includes loan-related information, such as loan money information, borrower information and the like, and the user is reminded of the existence of the abnormal loan through the first anti-fraud reminding information. Therefore, the anti-fraud capacity of the bank can be improved, the identification accuracy of the abnormal user loan is improved, and better defense can be dealt with.
Further, as shown in fig. 4, the constructing a user knowledge graph after extracting knowledge from the unstructured data set and the semi-structured data set and performing knowledge fusion with the structured data set includes:
s221: performing data integration on the structured data set;
s222: performing entity alignment on the unstructured data set and the semi-structured data set after knowledge extraction and the integrated structured data set;
s223: through knowledge reasoning, ontology construction is carried out based on data after entity alignment;
s224: and performing quality evaluation on the constructed ontology, updating knowledge, and constructing the user knowledge graph after the quality evaluation is passed.
Specifically, the knowledge graph is a structured semantic knowledge base, basic composition units are entity-relation-entity triples, the knowledge graph is a network knowledge base formed by linking entities with attributes through relations, and people can be relieved from a mode of manually filtering web pages to find answers through structured knowledge which is classified and sorted.
And carrying out structured classification on the multi-party data information to obtain an unstructured data set, a semi-structured data set and a structured data set. And performing data integration on the structured data, including but not limited to integration on entities and connections of different structured data. The unstructured data and the semi-structured data usually occupy a larger proportion and contain more high-value data, and the extraction of the knowledge of the entities, the relations and the attributes of the unstructured data set and the semi-structured data set through knowledge extraction comprises entity extraction, relation extraction and attribute extraction. The knowledge extraction can be carried out by applying tools such as natural language processing, image processing, social network analysis, machine learning and the like through searching, filtering and calculating to obtain entities, relations and attributes of the unstructured data set and the semi-structured data set. To give an example without limitation: the data extraction is carried out on the semi-structured data by adopting a crawler technology, a wrapper and a regular expression, and the data extraction is carried out on the unstructured data such as text data by adopting a natural language processing technology.
And integrating after knowledge extraction, and performing entity alignment on the structured data set, wherein the entity alignment includes relationship alignment and attribute alignment besides aligning entities, and can be realized by technologies such as similarity calculation, aggregation, clustering and the like. And carrying out entity alignment on the unstructured data set and the semi-structured data set with the structured data set, namely a knowledge fusion process, and keeping the same entity. And performing knowledge inference according to the existing data model and data and inference rules, wherein the inference rules comprise but are not limited to logical relations between entities, generative rules, predicate rules and the like.
And constructing a user knowledge graph body according to the data after entity alignment and the relation between the entities, wherein the constructed knowledge graph may have some errors and needs to be subjected to quality evaluation, and the knowledge graph is subjected to completion and knowledge graph error detection, such as knowledge graph completion and error detection based on an inference rule method. And further updating knowledge, adopting an updating feedback mechanism to continuously evolve the constructed ontology, judging that the quality evaluation is passed when a set quality evaluation standard is reached, and constructing the user knowledge graph. A database containing mass data can be constructed for main bodies such as banks through constructing a user knowledge map, deep analysis of bank users is facilitated, and rapid and accurate fraud identification of the banks is facilitated.
In summary, the graph database technology-based anti-fraud method and system provided by the embodiment of the present application have the following technical effects:
1. due to the adoption of a multi-party data sharing mode, multi-party data information about a user set is obtained; using the multi-party data information to construct a user knowledge graph; further retrieving the user knowledge graph to obtain a retrieval data set; the method comprises the steps of carrying out deep analysis on the retrieval data set, carrying out clustering analysis after normalization processing, constructing a training data set according to clustering results, carrying out exception detection on loan behaviors by using an isolated forest method, and obtaining first anti-fraud reminding information according to exception detection results, namely exception loan information, so as to carry out user loan exception reminding on main bodies such as banks.
2. Due to the adoption of the normalization, cluster analysis and principal component analysis methods, the characteristic data of different dimensions can be normalized, clustered and dimension-reduced, so that the redundant data volume is eliminated under the condition of ensuring sufficient data volume, the sample volume of data operation is reduced, the loss after dimension reduction is ensured to be minimum, and the technical effect of improving the speed of data operation is further achieved.
Example two
Based on the same inventive concept as the graph database technology-based anti-fraud method in the foregoing embodiment, as shown in fig. 5, an embodiment of the present application provides an anti-fraud system based on graph database technology, wherein the system includes:
the first obtaining unit 11 is configured to obtain multi-party data information of a user set based on data sharing between an anti-fraud analysis platform and a multi-party data source platform;
a first construction unit 12, wherein the first construction unit 12 is configured to construct a user knowledge graph based on the multi-party data information;
a second obtaining unit 13, where the second obtaining unit 13 is configured to perform targeted retrieval on the user knowledge graph to obtain a retrieval data set;
the second construction unit 14 is configured to perform normalization processing on the search data set, perform cluster analysis, and construct a training data set according to a cluster result;
a third obtaining unit 15, where the third obtaining unit 15 is configured to perform abnormal detection on loan behaviors through an isolated forest method according to the training data set, to obtain abnormal loan information;
a fourth obtaining unit 16, where the fourth obtaining unit 16 is configured to obtain first anti-fraud prompting information based on the abnormal loan information, and prompt the user that the loan is abnormal through the first anti-fraud prompting information.
Further, the system comprises:
the system comprises a first setting unit, a second setting unit and a third setting unit, wherein the first setting unit is used for setting pre-calibrated information, and the pre-calibrated information comprises loan amount characteristics, money information, associated advertisement characteristics and character vulnerability characteristics, wherein direct or indirect relevance exists among the characteristics in the pre-calibrated information;
a fifth obtaining unit, configured to perform target retrieval through the user knowledge graph according to the pre-scaled information, and obtain the retrieval data set.
Further, the system comprises:
a sixth obtaining unit, configured to obtain a first search data set after performing normalization processing on the search data set;
a seventh obtaining unit, configured to perform cluster partitioning on the first search data set to obtain a first clustering result, where the first clustering result includes multiple data sets of different categories;
and the third construction unit is used for carrying out principal component analysis on the data sets of different classes to construct the training data set.
Further, the system comprises:
a first execution unit to randomly draw a set of test samples from the training data set without being placed back;
a second execution unit, configured to randomly extract first test data from the test sample set as a root node of an isolation tree;
an eighth obtaining unit, configured to obtain feature value information of the first test data;
a ninth obtaining unit, configured to perform binary division on the test sample set according to the eigenvalue information to obtain a left branch and a right branch, where a eigenvalue of sample data in the left branch is smaller than the eigenvalue information; the eigenvalue of the sample data in the right branch is greater than the eigenvalue information;
a third execution unit, configured to repeat the binary division on the branch generated by the root node until a predetermined condition is reached, stop the binary division, and use the isolated branch as the abnormal loan information.
Further, the system comprises:
a tenth obtaining unit, configured to obtain, based on data sharing between the anti-fraud analysis platform and the multi-party data source platform, multi-party data information of the user set, where the multi-party data information includes user social relationships, transaction pattern associations, internet behaviors, and mobile device data characteristics.
Further, the system comprises:
an eleventh obtaining unit, configured to perform structured classification on the multi-party data information, so as to obtain a structured data set, an unstructured data set, and a semi-structured data set;
and the fourth construction unit is used for performing knowledge fusion on the unstructured data set and the semi-structured data set after the knowledge extraction and the structured data set, and constructing the user knowledge graph.
Further, the system comprises:
a fourth execution unit to perform data integration on the structured data set;
a fifth execution unit, configured to perform entity alignment on the unstructured dataset and the semi-structured dataset after knowledge extraction and the integrated structured dataset;
the sixth execution unit is used for performing ontology construction based on the data after entity alignment through knowledge reasoning;
and the fifth construction unit is used for performing quality evaluation on the constructed body and updating knowledge, and when the quality evaluation passes, the user knowledge graph is constructed.
Exemplary electronic device
The electronic apparatus of the embodiment of the present application is described below with reference to fig. 6.
Based on the same inventive concept as the graph database technology-based anti-fraud method in the foregoing embodiments, the present application further provides an anti-fraud system based on the graph database technology, including: a processor coupled to a memory, the memory for storing a program that, when executed by the processor, causes a system to perform the method of any of the first aspects.
The electronic device 300 includes: processor 302, communication interface 303, memory 301. Optionally, the electronic device 300 may also include a bus architecture 304. Wherein, the communication interface 303, the processor 302 and the memory 301 may be connected to each other through a bus architecture 304; the bus architecture 304 may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus architecture 304 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in FIG. 6, but this is not intended to represent only one bus or type of bus.
The communication interface 303 may be any device, such as a transceiver, for communicating with other devices or communication networks, such as an ethernet, a Radio Access Network (RAN), a Wireless Local Area Network (WLAN), a wired access network, and the like.
The memory 301 may be, but is not limited to, a ROM or other type of static storage device that can store static information and instructions, a RAM or other type of dynamic storage device that can store information and instructions, an electrically erasable Programmable read-only memory (EEPROM), a compact disk read-only memory (CD-ROM) or other optical disk storage, optical disk storage (including compact disk, laser disk, optical disk, digital versatile disk, blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. The memory may be self-contained and coupled to the processor through a bus architecture 304. The memory may also be integral to the processor.
The memory 301 is used for storing computer-executable instructions for executing the present application, and is controlled by the processor 302 to execute. The processor 302 is configured to execute the computer-executable instructions stored in the memory 301, thereby implementing an anti-fraud method based on graph database technology according to the above-described embodiments of the present application.
Optionally, the computer-executable instructions in the embodiments of the present application may also be referred to as application program codes, which are not specifically limited in the embodiments of the present application.
The embodiment of the application provides an anti-fraud method based on a graph database technology, wherein the method comprises the following steps: obtaining multi-party data information related to a user set in a multi-party data sharing mode; using the multi-party data information to construct a user knowledge graph; further retrieving the user knowledge graph to obtain a retrieval data set; and carrying out deep analysis on the retrieval data set, carrying out clustering analysis after normalization processing, constructing a training data set according to clustering results, carrying out abnormal detection on loan behaviors by using an isolated forest method, and obtaining first anti-fraud reminding information according to abnormal detection results, namely abnormal loan information, so as to remind users of loan abnormity for main bodies such as banks.
Those of ordinary skill in the art will understand that: various numbers of the first, second, etc. mentioned in this application are only for convenience of description and distinction, and are not used to limit the scope of the embodiments of this application, nor to indicate a sequence order. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one" means one or more. At least two means two or more. "at least one," "any," or similar expressions refer to any combination of these items, including any combination of item(s) or item(s). For example, at least one (one ) of a, b, or c, may represent: a, b, c, a-b, a-c, b-c, or a-b-c, wherein a, b, c may be single or multiple.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, it may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the application to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device including one or more available media integrated servers, data centers, and the like. The usable medium may be a magnetic medium (e.g., a floppy Disk, a hard Disk, a magnetic tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a Solid State Disk (SSD)), among others.
The various illustrative logical units and circuits described in this application may be implemented or operated upon by design of a general purpose processor, a digital signal processor, an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof. A general-purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, e.g., a digital signal processor and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a digital signal processor core, or any other similar configuration.
The steps of a method or algorithm described in the embodiments herein may be embodied directly in hardware, in a software element executed by a processor, or in a combination of the two. The software cells may be stored in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. For example, a storage medium may be coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC, which may be disposed in a terminal. In the alternative, the processor and the storage medium may reside in different components within the terminal. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Although the present application has been described in conjunction with specific features and embodiments thereof, it will be evident that various modifications and combinations can be made thereto without departing from the spirit and scope of the application. Accordingly, the specification and figures are merely exemplary of the application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of the application. It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the present application and its equivalent technology, it is intended that the present application include such modifications and variations.
Claims (6)
1. An anti-fraud method based on graph database technology, the method being applied to an anti-fraud analysis platform, the platform being communicatively connected to a multi-party data source platform, the method comprising:
obtaining multi-party data information of a user set based on data sharing of the anti-fraud analysis platform and the multi-party data source platform;
constructing a user knowledge graph based on the multi-party data information;
performing targeted retrieval on the user knowledge graph to obtain a retrieval data set;
carrying out normalization processing on the retrieval data set, then carrying out clustering analysis, and constructing a training data set according to a clustering result;
performing abnormal detection on loan behaviors through an isolated forest method according to the training data set to obtain abnormal loan information;
obtaining first anti-fraud reminding information based on the abnormal loan information, and reminding the user of abnormal loan through the first anti-fraud reminding information;
the step of performing targeted retrieval on the user knowledge graph to obtain a retrieval data set comprises the following steps:
setting pre-calibrated information, wherein the pre-calibrated information comprises loan line characteristics, money information, associated advertisement characteristics and character vulnerability characteristics, and direct or indirect relevance exists among the characteristics in the pre-calibrated information;
according to the pre-calibrated information, performing target retrieval through the user knowledge graph to obtain the retrieval data set;
the step of performing normalization processing on the retrieval data set, then performing clustering analysis, and constructing a training data set according to a clustering result comprises the following steps:
after normalization processing is carried out on the retrieval data set, a first retrieval data set is obtained;
clustering and dividing the first retrieval data set to obtain a first clustering result, wherein the first clustering result comprises a plurality of data sets of different categories;
performing principal component analysis on the data sets of different categories to construct the training data set;
the method comprises the following steps of carrying out abnormal detection on loan behaviors through an isolated forest method according to the training data set to obtain abnormal loan information, and comprises the following steps:
randomly non-replaced drawn test sample sets from the training data set;
randomly extracting first test data from the test sample set as a root node of an isolation tree;
obtaining characteristic value information of the first test data;
performing binary division on the test sample set according to the characteristic value information to obtain a left branch and a right branch, wherein the characteristic value of sample data in the left branch is smaller than the characteristic value information; the eigenvalue of the sample data in the right branch is greater than the eigenvalue information;
repeating the binary division on the branches generated by the root node until a predetermined condition is reached, stopping the binary division, and taking the isolated branches as the abnormal loan information.
2. The method of claim 1, wherein the obtaining of the multi-party data information of the user set based on the data sharing of the anti-fraud analysis platform and the multi-party data source platform comprises:
and acquiring multi-party data information of the user set based on the data sharing of the anti-fraud analysis platform and the multi-party data source platform, wherein the multi-party data information comprises user social relations, transaction mode association, internet behaviors and mobile equipment data characteristics.
3. The method of claim 1, wherein said constructing a user knowledge graph based on said multi-party data information comprises:
carrying out structured classification on the multi-party data information to obtain a structured data set, an unstructured data set and a semi-structured data set;
and extracting knowledge from the unstructured data set and the semi-structured data set, and then performing knowledge fusion on the unstructured data set and the structured data set to construct the user knowledge graph.
4. The method of claim 3, wherein said constructing the user knowledge graph after knowledge fusion of the unstructured dataset and the semi-structured dataset with the structured dataset after knowledge extraction comprises:
performing data integration on the structured data set;
performing entity alignment on the unstructured data set and the semi-structured data set after knowledge extraction and the integrated structured data set;
performing ontology construction based on the data after entity alignment through knowledge reasoning;
and performing quality evaluation on the constructed ontology, updating knowledge, and constructing the user knowledge graph after the quality evaluation is passed.
5. A graph database technology based anti-fraud system, characterized in that said system comprises:
the first obtaining unit is used for obtaining multi-party data information of a user set based on data sharing of the anti-fraud analysis platform and the multi-party data source platform;
a first construction unit for constructing a user knowledge graph based on the multi-party data information;
a second obtaining unit, configured to perform targeted retrieval on the user knowledge graph to obtain a retrieval data set;
the second construction unit is used for carrying out normalization processing on the retrieval data set, then carrying out clustering analysis, and constructing a training data set according to a clustering result;
the third obtaining unit is used for carrying out abnormal detection on loan behaviors through an isolated forest method according to the training data set to obtain abnormal loan information;
a fourth obtaining unit, configured to obtain first anti-fraud prompting information based on the abnormal loan information, and prompt the user that the loan is abnormal through the first anti-fraud prompting information;
the system comprises a first setting unit, a second setting unit and a third setting unit, wherein the first setting unit is used for setting pre-calibrated information, and the pre-calibrated information comprises loan amount characteristics, money information, associated advertisement characteristics and character vulnerability characteristics, wherein direct or indirect relevance exists among the characteristics in the pre-calibrated information;
a fifth obtaining unit, configured to perform target retrieval through the user knowledge graph according to the pre-calibrated information, and obtain the retrieval data set;
a sixth obtaining unit, configured to obtain a first search data set after performing normalization processing on the search data set;
a seventh obtaining unit, configured to perform cluster partitioning on the first search data set to obtain a first clustering result, where the first clustering result includes multiple data sets of different categories;
a third construction unit, configured to perform principal component analysis on the data sets of different categories to construct the training data set;
a first execution unit to randomly draw a set of test samples without playback from the training data set;
a second execution unit, configured to randomly extract first test data from the test sample set as a root node of an isolation tree;
an eighth obtaining unit, configured to obtain feature value information of the first test data;
a ninth obtaining unit, configured to perform binary division on the test sample set according to the eigenvalue information to obtain a left branch and a right branch, where an eigenvalue of sample data in the left branch is smaller than the eigenvalue information; the eigenvalue of the sample data in the right branch is greater than the eigenvalue information;
a third execution unit, configured to repeat the binary division on the branch generated by the root node until a predetermined condition is reached, stop the binary division, and use the isolated branch as the abnormal loan information.
6. An anti-fraud system based on graph database technology, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method according to any of claims 1-4 are implemented when the processor executes the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111609100.0A CN113988878B (en) | 2021-12-27 | 2021-12-27 | Graph database technology-based anti-fraud method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111609100.0A CN113988878B (en) | 2021-12-27 | 2021-12-27 | Graph database technology-based anti-fraud method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113988878A CN113988878A (en) | 2022-01-28 |
| CN113988878B true CN113988878B (en) | 2022-07-19 |
Family
ID=79734505
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111609100.0A Active CN113988878B (en) | 2021-12-27 | 2021-12-27 | Graph database technology-based anti-fraud method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113988878B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114897210B (en) * | 2022-03-30 | 2023-09-12 | 南通电力设计院有限公司 | Energy source interconnection method and system based on energy source loss prediction |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109816397A (en) * | 2018-12-03 | 2019-05-28 | 北京奇艺世纪科技有限公司 | A kind of fraud method of discrimination, device and storage medium |
| CN110765117A (en) * | 2019-09-30 | 2020-02-07 | 中国建设银行股份有限公司 | Fraud identification method and device, electronic equipment and computer-readable storage medium |
| CN111833172A (en) * | 2020-05-25 | 2020-10-27 | 百维金科(上海)信息科技有限公司 | Consumption credit fraud detection method and system based on isolated forest |
| CN113553446A (en) * | 2021-07-28 | 2021-10-26 | 厦门国际银行股份有限公司 | Financial anti-fraud method and device based on heteromorphic graph deconstruction |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112053221A (en) * | 2020-08-14 | 2020-12-08 | 百维金科(上海)信息科技有限公司 | Knowledge graph-based internet financial group fraud detection method |
-
2021
- 2021-12-27 CN CN202111609100.0A patent/CN113988878B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109816397A (en) * | 2018-12-03 | 2019-05-28 | 北京奇艺世纪科技有限公司 | A kind of fraud method of discrimination, device and storage medium |
| CN110765117A (en) * | 2019-09-30 | 2020-02-07 | 中国建设银行股份有限公司 | Fraud identification method and device, electronic equipment and computer-readable storage medium |
| CN111833172A (en) * | 2020-05-25 | 2020-10-27 | 百维金科(上海)信息科技有限公司 | Consumption credit fraud detection method and system based on isolated forest |
| CN113553446A (en) * | 2021-07-28 | 2021-10-26 | 厦门国际银行股份有限公司 | Financial anti-fraud method and device based on heteromorphic graph deconstruction |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113988878A (en) | 2022-01-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rao et al. | Design of comprehensive evaluation index system for P2P credit risk of “three rural” borrowers | |
| CN111767716B (en) | Method and device for determining enterprise multi-level industry information and computer equipment | |
| US20140143186A1 (en) | Hybrid clustering for data analytics | |
| CN110929525B (en) | Network loan risk behavior analysis and detection method, device, equipment and storage medium | |
| CN112287020B (en) | Big data mining method based on graph analysis | |
| CN108090216A (en) | Label prediction method, device and storage medium | |
| CN112132238A (en) | Method, device, equipment and readable medium for identifying private data | |
| CN114792246B (en) | Product typical feature mining method and system based on topic integrated clustering | |
| CN112232944A (en) | Scoring card creating method and device and electronic equipment | |
| CN113988878B (en) | Graph database technology-based anti-fraud method and system | |
| Wei et al. | Online education recommendation model based on user behavior data analysis | |
| CN113378090A (en) | Internet website similarity analysis method and device and readable storage medium | |
| Zhao et al. | Topic identification of text‐based expert stock comments using multi‐level information fusion | |
| CN112434126B (en) | Information processing method, device, equipment and storage medium | |
| CN112801784A (en) | Bit currency address mining method and device for digital currency exchange | |
| CN116701658A (en) | Business intention determining method and device and computer equipment | |
| CN116933130A (en) | Enterprise industry classification method, system, equipment and medium based on big data | |
| CN116205296A (en) | ABAC strategy engineering method integrating top-down and bottom-up | |
| CN116108847A (en) | Knowledge graph construction method, CWE community description method and storage medium | |
| CN115640378A (en) | Work order retrieval method, server, medium and product | |
| CN115496066A (en) | Text analysis system, text analysis method, electronic device, and storage medium | |
| CN115310606A (en) | Deep learning model depolarization method and device based on data set sensitive attribute reconstruction | |
| CN114722941A (en) | Credit default identification method, apparatus, device and medium | |
| CN114529399A (en) | User data processing method, device, computer equipment and storage medium | |
| CN113987536A (en) | Method and device for determining security level of field in data table, electronic equipment and medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20221103 Address after: F11, Building G, Yunmi City, No. 19-11, Ningshuang Road, Yuhuatai District, Nanjing City, Jiangsu Province, 210000 Patentee after: Jiangsu Zhiqi Cloud Big Data Technology Co.,Ltd. Address before: 210000 floor 11, building g, yunmi City, No. 19, ningshuang Road, Yuhuatai District, Nanjing, Jiangsu Province Patentee before: COGNITIVE COMPUTING NANJING INFORMATION TECHNOLOGY Co.,Ltd. |
|
| TR01 | Transfer of patent right |