Gang detection method based on financial transaction network and implementation device thereof
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A financial transaction and detection method technology, applied in finance, neural learning methods, biological neural network models, etc., can solve problems such as high misclassification rate, inability to separate, and unrecognizable members
Pending Publication Date: 2020-10-30
HARBIN INST OF TECH AT WEIHAI +1
View PDF2 Cites 3 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
However, in this patent, 1) the biggest problem in group identification based on the degree of cohesion is that the obtained groups are often closely adjacent members on the graph, and non-adjacent members cannot be identified; 2) the inherent problem of the algorithm, two nodes in the agglomeration method After merging, it cannot be separated, so the misclassification rate is high; 3) The use of transaction sequences stays on simple topological relationships, and the results are unreliable
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0087] A gang detection method based on financial transaction network, such as figure 1 and Figure 4 shown, including:
[0088] (1) Data preprocessing: data cleaning of transaction data, extraction of each user's transaction sequence, and construction of graph data;
[0089] In the described step (1), the transaction data includes user, counterparty account number, transaction time and transaction amount, and the specific steps of data cleaning to the transaction data include:
[0090] 1-1. Missing value filling: If any of the fields of the user, counterparty account, and transaction time of a certain transaction data is missing, the transaction data will be discarded;
[0091] If only the transaction amount field in a certain transaction data is missing, the average value filling method is used to fill it, that is, the average value of all transaction amounts of the current user is calculated, and the average value is used to fill the transaction amount;
[0092] 1-2. Dat...
Embodiment 2
[0149] The implementation device based on the gang detection method of the financial transaction network provided in embodiment 1 includes:
[0150] The data preprocessing module is used to clean the transaction data, extract the transaction sequence of each user, construct the graph data, and execute step (1);
[0151] The user feature vector generation module uses the sequence model to obtain the user time-series feature vector, uses the GAE model to obtain the user space feature vector, and normalizes the user time-series feature vector and the space feature vector respectively, and connects them to perform step (2) ;
[0152] The gang detection module is used to calculate the gang to which each node belongs, and output the gang mark of the node, which is used to execute step (3).
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
PUM
Login to view more
Abstract
The invention relates to a gang detection method based on a financial transaction network and an implementation device thereof. The detection method comprises the following steps: (1) data preprocessing; (2) generating of a user feature vector: obtaining a user time sequence feature vector by using a sequence model, and obtaining a user space feature vector by using a GAE model; normalizing the user time sequence feature vector and the space feature vector respectively, and performing connection operation to generate a node representation vector; (3) gang detection: calculating a gang to whicheach node belongs, and outputting gang marks of the nodes. According to the method, original financial transaction flow information data is utilized, time sequence features and space structure features are extracted firstly, then connection features are used for calculating the distance between every two nodes to serve as a weight, and each user can be allocated to a potential gang by using a gang detection algorithm based on modularity optimization.
Description
technical field [0001] The invention relates to a gang detection method based on a financial transaction network and an implementation device thereof, belonging to the technical field of data mining. Background technique [0002] Group detection refers to the detection of node sets with the same characteristics on graph data, and it is also called community detection in the field of complex networks. Gang detection has a broad application base. Aided decision-making tools in financial crime require high accuracy and interpretability. Therefore, mining potential suspects in the massive transaction flow has extensive research and application value. At present, this work generally relies on manual data mining and analysis, which requires a deep understanding of data and criminal behavior and in-depth analysis of data, and requires high human experience. With the large amount of transaction data The outbreak has posed new and huge challenges to both machine hardware and peopl...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
Application Information
Patent Timeline
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to view more 
Login to view more