A three-dimensional visualization system based on Ethereum big data and a construction method thereof

Abstract

The application discloses a three-dimensional visualization system and construction method based on Ethereum big data, which classifies Ethereum blockchain big data according to entities, and constructs an Ethereum head ecological network model by using a Spark computing framework. The whole Ethereum network model is laid out by using a three-dimensional force guiding model, and the Barnes-Hut optimization algorithm extended to three dimensions is introduced in the graph layout process. The local part is laid out on the surface by using a spherical layout method. Finally, the WebGL, ThreeJS and other WEB three-dimensional graphics rendering technologies are used to realize the visualization display of the Ethereum head ecological network on the WEB side. The user can view the three-dimensional visualization model of the Ethereum transaction network in different ways, so that the user interaction process becomes more smooth, the details of the three-dimensional visualization model are controlled more deeply, and the search function in the model is provided, so that the user can quickly locate the node to be concerned.

Description

Technical Field

[0001] This invention belongs to the field of data visualization processing technology, specifically relating to a web 3D visualization system and method based on Ethereum big data. Background Technology

[0002] Blockchain technology emerged from the concept of decentralized infrastructure and distributed storage consensus technology. Based on blockchain technology, engineers implemented a Turing-complete Ethereum Virtual Machine (EVM). By running smart contracts on the EVM, a decentralized blockchain platform was built. Ethereum has also rapidly become a leading public blockchain due to its ease of use, faster transaction speeds, and more complete ecosystem.

[0003] After years of development, applications based on blockchain technology have expanded to areas such as digital cryptocurrencies, supply chain management, copyright protection, and healthcare. Blockchain essentially addresses issues of privacy and security, information traceability, transaction compliance, data authenticity, and process efficiency. However, with the widespread application of blockchain technology, a series of blockchain transaction security issues have arisen and become increasingly prominent. Therefore, there is an urgent need for appropriate analytical tools to address the complexities of blockchain transactions. Blockchain visualization can provide exploratory insights into observable potential behaviors within on-chain data. Because 3D data visualization can present data characteristics in an intuitive, three-dimensional, and multi-faceted way, 3D visualization of blockchain transaction data can display the entire transaction activity in a top-down manner, thereby obtaining the data structure characteristics of such recurring abnormal behaviors.

[0004] However, given the current business needs for Ethereum transaction data visualization, most Ethereum transaction data visualizations are based on basic charts, such as bar charts, pie charts, histograms, scatter plots, and heatmaps. A small number are based on the transaction network to show the transaction relationships between Ethereum accounts. But there is still no 3D visualization of Ethereum transaction data to comprehensively show the relationships between Ethereum users, smart contracts, and DApps. Summary of the Invention

[0005] To address the shortcomings of current technology and practical needs, this application provides a three-dimensional visualization construction method based on Ethereum big data. It aims to solve user needs and facilitate data analysis for relevant personnel, providing more dimensional visualization effects for Ethereum big data.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a three-dimensional visualization construction method based on Ethereum big data, comprising the following steps:

[0007] Collect synchronized Ethereum data and parse and process it to obtain external transaction information tables, account balance tables, and token balance tables. Use web crawling technology to obtain DApp information tables.

[0008] Further clean the DApp information table to obtain the top Ethereum transaction network within a set time period;

[0009] Based on the Ethereum head transaction network, an optimized force-directed layout algorithm and a local spherical layout algorithm are used to construct a three-dimensional Ethereum transaction network model;

[0010] Using Vue to build a visual website, the 3D network model data after the Ethereum transaction network layout is transmitted to a 3D graphics rendering engine, the nodes and the relationships between the nodes are drawn, the final model is rendered, and the 3D interaction methods are defined, providing two perspective interaction modes: track controller and third-person view controller.

[0011] When cleaning the DApp information table to obtain the top Ethereum transaction network for the past month, the Spark large-scale data processing engine was used to count the number of DApp users, the total number of external transactions between accounts, and the number of times smart contracts were used within the month. Statistical distribution histograms were then plotted. Based on the inflection points of each distribution chart, thresholds were determined for the number of DApp users, the amount of external transactions, and the number of times smart contracts were used. Based on these thresholds, external transactions with transaction amounts exceeding the threshold, smart contracts with transaction counts exceeding the threshold, and DApps with more users than the threshold were filtered to obtain the top Ethereum transaction network for the past month. The visualization results are shown in the reference. Figure 1 .

[0012] The optimization process of the force-guided layout algorithm is as follows:

[0013] First, based on the node attributes in the Ethereum transaction network, nodes are generally divided into three categories: EOA account nodes, smart contract nodes, and token nodes. Smart contract nodes and token nodes are further subdivided according to the type of DApp they belong to: games, finance, social, and exchanges. Based on the relationship attributes in the Ethereum transaction network, edges are generally divided into five categories: external transaction relationships, user-invoiced smart contract relationships, user-held token relationships, smart contract and DApp ownership relationships, and token and DApp ownership relationships.

[0014] Secondly, the coordinates of all nodes in three-dimensional space are randomly initialized; the nodes are treated as electrons, so that there is a repulsive force between all nodes, and the relationship is treated as a spring, so that there is an attractive force between the nodes in the relationship.

[0015] Then, the formulas for calculating attraction and repulsion are modified to reduce the repulsion between smart contracts and tokens belonging to the same DApp and increase the attraction between smart contracts and tokens belonging to the same DApp. At the same time, the formula for calculating the central force is modified to make the entire network layout tighten towards the center and to use the Barnes-Hut tree structure to calculate the repulsion between nodes.

[0016] Under the combined action of the repulsive and attractive forces, the node coordinates are gradually adjusted to construct a three-dimensional network topology.

[0017] The three-dimensional network topology iterates multiple times under the action of repulsive force, attraction force and central force, and finally reaches an equilibrium state. The coordinates of all nodes in the three-dimensional space are basically fixed, and the stable three-dimensional space coordinates of all nodes are obtained. The coordinates of each DAPP in the three-dimensional space are calculated according to the smart contract and token position of the DAPP.

[0018] The three-dimensional spatial coordinates of users and tokens of the DAPP and the smart contract points of the DAPP are determined by the spherical layout method and the in-sphere layout method, respectively, and the three-dimensional spatial stable coordinates of all nodes in the Ethereum transaction network are obtained.

[0019] The calculations for attraction, repulsion, and central force are as follows:

[0020] a. The formula for calculating gravity is:

[0021] b. The formula for calculating repulsive force is:

[0022] c. The formula for calculating the central force is:

[0023] Using an octree data structure, the network model in the initialized 3D space is recursively divided into 8 sub-regions. If a sub-region does not contain any nodes, it is not processed in the subsequent recursion. If a region contains only one node, it becomes a leaf node in the octree. If a region contains two or more nodes, it is further divided according to the above principle. Finally, all sub-regions construct a non-complete octree. Each node in the tree has at most 8 child nodes, and the leaf nodes in the tree have exactly one node in the Ethereum transaction 3D network.

[0024] Based on the Barnes-Hut algorithm, nearby objects are grouped and treated as a whole. If the distance between objects is large enough, the effect of gravity is approximated by the centroid. The centroid of a group of objects is the average position of all objects in the group, and the mass of the objects is used as the weight. If the positions of two objects are ( x 1, y 1,z 1) and ( x 2, y 2, z 2), the masses are respectively m 1 and m 2. Its total mass and total centroid are respectively:

[0025] Total mass:

[0026] Overall centroid: .

[0027] The Barnes-Hut tree is constructed as follows:

[0028] If the (root) node x does not contain an object, then place the new node b into it.

[0029] If the (root) node x is an internal node (i.e., a non-leaf node), update the total mass and centroid of x, and recursively insert node b into one of the eight regions.

[0030] If the (root) node x is an external node (i.e. a leaf node) containing a node c, then that means there are two nodes b and c in the same region. The region is further divided into eight sub-regions, and then nodes b and c are recursively inserted into the corresponding branches. Finally, the centroid and total mass of node x are updated.

[0031] The repulsive forces between nodes are calculated using the constructed Barnes-Hut tree structure. The specific steps are as follows:

[0032] ① If the current node is a leaf node and the node is not b, calculate the force it exerts on b and add the result to the net force on b.

[0033] ② Otherwise, calculate the ratio s / d. If s / d0 < θ, s is the width of the region where the internal node is located, d is the distance between the object and the centroid of the node, and θ is the threshold, set to 0.5. Treat the internal node as a node and calculate the force exerted by the node on b. Add the result to the net force of b.

[0034] ③ Otherwise, recursively run steps ① and ② on the child nodes of this node.

[0035] Additionally, a 3D visualization system based on Ethereum big data is provided, including an Ethereum transaction data acquisition and storage module, a data cleaning module, a 3D visualization model construction module, and a 3D visualization model display module;

[0036] The Ethereum transaction data storage module is used to break down the Ethereum transaction big data obtained from the OpenEthereum Ethereum third-party wallet client into different themes and primary keys using EthereumETL tools, and then store it in the HDFS distributed file system under the Hadoop ecosystem for persistent storage. It also uses web crawling technology to obtain DApp data information from multiple websites and stores it in the relational database MySQL for persistent storage.

[0037] The data cleaning module utilizes the Spark large-scale data processing computing engine to filter out external transaction data with transaction amounts exceeding a threshold, smart contracts with transaction counts exceeding a threshold, and DApps with a number of users exceeding a threshold within a month from the external transaction information table, account balance table, token balance table, and DApp information table in the Ethereum transaction data, thereby obtaining the top Ethereum transaction networks within a month.

[0038] The 3D visualization model building module is based on the data results processed by the data cleaning module. It adopts a heuristic layout algorithm, namely the force-oriented layout algorithm, and optimizes it with the octree algorithm to build an initial 3D transaction network of the full data. Then, the spherical layout method is used for local areas to place the unique users of DAPP on its surface to obtain the Ethereum transaction 3D network model of the full data.

[0039] The 3D visualization model display module is based on the Ethereum transaction 3D network model generated by the 3D visualization model building module, and is drawn and rendered using ThreeJS and / or WebGL web-based 3D visualization technology.

[0040] The present invention also provides a computer device, including a processor and a memory, the memory being used to store a computer executable program, the processor reading the computer executable program from the memory and executing it, and the processor executing the computer executable program can realize the three-dimensional visualization construction method based on Ethereum big data.

[0041] Compared with the prior art, the present invention has at least the following beneficial effects:

[0042] Compared with existing technologies, the present invention proposes a three-dimensional visualization construction method based on Ethereum big data. It uses Ethereum transaction data pulled from the OpenEthereum Ethereum third-party wallet client and persistently stores it using the HDFS distributed file system under the Hadoop ecosystem. It also uses web crawling technology to obtain DApp data information from multiple sources and stores it in the relational database MySQL for persistent storage, which can obtain more comprehensive Ethereum-related data.

[0043] This invention uses the Spark large-scale data processing computing engine to filter out external transaction data, smart contracts with transaction counts exceeding a threshold, and DApps with user counts exceeding a threshold within a month from the external transaction information table, account balance table, token balance table, and DApp information table in Ethereum transaction data. This results in the top Ethereum transaction networks within a month, allowing for a more precise selection of the data visualization scope and a more comprehensive display of Ethereum transaction network information.

[0044] The Ethereum transaction network obtained after data cleaning is constructed using a heuristic layout algorithm, namely the force-directed layout algorithm, and optimized using the octree algorithm to build an initial three-dimensional transaction network of the full data. Then, the spherical layout method is used locally to place the unique users of DApps on its surface, thus obtaining a three-dimensional Ethereum transaction network model of the full data.

[0045] The calculation time for 3D graph layout has been further optimized, reducing the time complexity of the repulsion calculation process to log8N; at the same time, a new layout presentation mode that reflects the relationship between DAPP, smart contracts, and users is proposed.

[0046] This invention proposes to present a 3D visualization model of the Ethereum transaction network on a webpage using rendering technologies such as WebGL and ThreeJS. This presents significant challenges to browser response speed and rendering pressure. To address the issue of excessive time required for large-scale network deployment, a monthly scheduled calculation and 3D spatial layout of the Ethereum transaction network is employed. The resulting 3D network model is stored in in-memory Redis and periodically persisted to disk to prevent data loss due to server downtime. Front-end requests for model data are directly retrieved from the Redis database, significantly improving browser response speed. Furthermore, 3D models of nodes with the same attributes are merged to minimize draw calls, fully utilize GPU computing resources, and substantially improve rendering performance. Two interactive perspectives—a track controller and a third-person view controller—are provided, allowing users to view the 3D visualization model of the Ethereum transaction network in different ways. This makes the user interaction process more streamlined, provides deeper control over the details of the 3D visualization model, and offers an in-model search function, enabling users to quickly locate nodes of interest. Attached Figure Description

[0047] Figure 1 This is a schematic diagram illustrating a visualization result of the present invention. Detailed Implementation

[0048] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0049] This invention provides a three-dimensional visualization method based on Ethereum big data. It uses Ethereum transaction data pulled from an OpenEthereum Ethereum third-party wallet client and persistently stores it using the HDFS distributed file system under the Hadoop ecosystem. It also uses web crawling technology to obtain DApp data information from multiple sources and stores it in a relational database MySQL for persistent storage, thus obtaining a basic information table of Ethereum big data.

[0050] Data cleaning based on the basic information table yielded the top Ethereum transaction networks for the past month. Using the Spark large-scale data processing engine, the number of DApp users, the total number of external transactions between accounts, and the number of times smart contracts were used within the past month were statistically analyzed. A histogram of the statistical distribution was plotted, revealing that the results all belong to a long-tail distribution.

[0051] Based on the inflection points of each distribution map, thresholds for the number of DApp users, external transaction amount, and number of times smart contracts are used are determined. Based on these thresholds, external transaction data with transaction amount exceeding the threshold, smart contracts with transaction frequency exceeding the threshold, and DApps with more users than the threshold are filtered out within a month to obtain the top Ethereum transaction network within a month.

[0052] An optimized force-directed layout algorithm and a local spherical layout algorithm are used to construct a three-dimensional Ethereum transaction network. First, nodes are broadly categorized into three types based on their attributes: EOA account nodes, smart contract nodes, and token nodes. Smart contract nodes and token nodes are further subdivided into categories such as games, finance, social networking, and exchanges based on their respective DApps. Edges are broadly categorized into five types based on their relationship attributes: external transaction relationships, user-invoiced smart contract relationships, user-held token relationships, smart contract-DApp ownership relationships, and token-DApp ownership relationships. Second, the coordinates of all nodes in three-dimensional space are randomly initialized. The nodes are treated as electrons, creating repulsive forces between them; the relationships are treated as springs, creating attractive forces between nodes within those relationships. During this process, the formulas for calculating these forces are modified to decrease the repulsive forces between smart contracts and tokens belonging to the same DApp and increase the attractive forces between them. Simultaneously, the formula for calculating the central force is modified to tighten the network layout towards the center. The NP-hard problem in repulsion calculation is solved by constructing a Barnes-Hut tree using the key idea of ​​the Barnes-Hut algorithm. The specific steps are as follows:

[0053] 1. If the (root) node x does not contain an object, then place the new node b into it.

[0054] 2. If the (root) node x is an internal node (i.e., a non-leaf node), update the total mass and centroid of x. Recursively insert node b into one of the eight regions.

[0055] 3. If the (root) node x is an external node (i.e., a leaf node) and it contains a node c, then this means that there are two nodes b and c in the same region. Therefore, this region is further divided into eight sub-regions. Then, nodes b and c are recursively inserted into their corresponding branches; because b and c may still be in the same sub-region, a single insertion operation may involve multiple sub-region partitions, ultimately updating the centroid and total mass of node x.

[0056] The repulsive forces between nodes are calculated using the constructed Barnes-Hut tree structure. The specific steps are as follows:

[0057] 1. If the current node is a leaf node (and the node is not b), calculate the force it exerts on b and add the result to the net force on b.

[0058] 2. Otherwise, calculate the ratio s / d. If s / d0 < θ, where s is the width of the region where the internal node is located, d is the distance between the object and the centroid of the node, and θ is a threshold set to 0.5, treat the internal node as a node, calculate the force exerted by the node on b, and add the result to the net force of b.

[0059] 3. Otherwise, run this program recursively on the child nodes of this node.

[0060] Under the combined action of the repulsive and attractive forces, the node coordinates are gradually adjusted to form a three-dimensional network topology.

[0061] The three-dimensional network topology iterates multiple times under the influence of repulsive, attractive, and central forces, eventually reaching an equilibrium state. The coordinates of all nodes in the three-dimensional space are basically fixed, thus obtaining the stable three-dimensional spatial coordinates of all nodes. Then, based on the smart contract and token positions of the DAPP, the coordinates of each DAPP in the three-dimensional space are calculated. The three-dimensional spatial coordinates of the users and tokens of the DAPP and the smart contract points of the DAPP are determined according to the spherical layout method and the in-sphere layout method, respectively. Finally, the stable three-dimensional spatial coordinates of all nodes in the Ethereum transaction network are obtained.

[0062] Using Vue to build a visualization website, the 3D network model data of the Ethereum transaction network layout is transmitted to a 3D graphics rendering engine. Nodes and the relationships between them are drawn, and the final model is rendered. Simultaneously, 3D interaction methods are defined, providing two perspective interaction modes: a track controller and a third-person view controller. This allows users to view the 3D visualization model of the Ethereum transaction network in different ways, making the user interaction process more streamlined and allowing for deeper control over the details of the 3D visualization model. Furthermore, a month's worth of Ethereum transaction network model files are stored in a Redis in-memory database, reducing model page loading speed, improving user experience, and further alleviating the pressure of 3D rendering.

[0063] In addition, the present invention can also provide a computer device, including a processor and a memory, wherein the memory is used to store a computer executable program, the processor reads part or all of the computer executable program from the memory and executes it, and the processor can realize the three-dimensional visualization construction method based on Ethereum big data described in the present invention when executing part or all of the computer executable program.

[0064] The computer device may be a laptop, a desktop computer, or a workstation.

[0065] The processor can be a central processing unit (CPU), a graphics processing unit (GPU) / digital signal processor (DSP), an application-specific integrated circuit (ASIC), or an off-the-shelf programmable gate array (FPGA).

[0066] The memory described in this invention can be an internal storage unit of a laptop, desktop computer, or workstation, such as memory or hard disk; or it can be an external storage unit, such as a portable hard disk or flash memory card.

[0067] The 3D visualization of Ethereum transaction data provides data mining researchers with insights into transaction behavior, wealth distribution, transaction generation mechanisms, and the identification of fluctuations in the cryptocurrency financial market. Given the large volume and rapid growth of Ethereum data, this invention utilizes big data processing techniques and graph layout algorithms to create a more rational data structure, presenting Ethereum transaction data in a 3D visualization that provides valuable reference for Ethereum and blockchain analysts.

Claims

1. A method for constructing 3D visualization based on Ethereum big data, characterized in that, Includes the following steps: Collect synchronized Ethereum data and parse and process it to obtain external transaction information tables, account balance tables, and token balance tables. Use web crawling technology to obtain DApp information tables. Further clean the DApp information table to obtain the top Ethereum transaction network within a set time period; Based on the Ethereum head transaction network, an optimized force-directed layout algorithm and a local spherical layout algorithm are used to construct a three-dimensional Ethereum transaction network model; the optimization process of the force-directed layout algorithm is as follows: First, based on the node attributes in the Ethereum transaction network, nodes are generally divided into three categories: EOA account nodes, smart contract nodes, and token nodes. Smart contract nodes and token nodes are further subdivided according to the type of DApp they belong to: games, finance, social, and exchanges. Based on the relationship attributes in the Ethereum transaction network, edges are generally divided into five categories: external transaction relationships, user-invoiced smart contract relationships, user-held token relationships, smart contract and DApp ownership relationships, and token and DApp ownership relationships. Secondly, the coordinates of all nodes in three-dimensional space are randomly initialized; the nodes are treated as electrons, so that there is a repulsive force between all nodes, and the relationship is treated as a spring, so that there is an attractive force between the nodes in the relationship. Then, the formulas for calculating attraction and repulsion are modified to reduce the repulsion between smart contracts and tokens belonging to the same DApp and increase the attraction between smart contracts and tokens belonging to the same DApp. At the same time, the formula for calculating the central force is modified to make the entire network layout tighten towards the center and to use the Barnes-Hut tree structure to calculate the repulsion between nodes. Under the combined action of the repulsive and attractive forces, the node coordinates are gradually adjusted to construct a three-dimensional network topology. The three-dimensional network topology iterates multiple times under the action of repulsive force, attractive force and central force, and finally reaches an equilibrium state. The coordinates of all nodes in the three-dimensional space are basically fixed, and the stable three-dimensional space coordinates of all nodes are obtained. The coordinates of each DAPP in the three-dimensional space are calculated according to the smart contract and token position of the DAPP. Using Vue to build a visual website, the 3D network model data after the Ethereum transaction network layout is transmitted to a 3D graphics rendering engine, the nodes and the relationships between the nodes are drawn, the final model is rendered, and the 3D interaction methods are defined, providing two perspective interaction modes: track controller and third-person view controller.

2. The method for constructing a 3D visualization based on Ethereum big data according to claim 1, characterized in that, When cleaning the DApp information table to obtain the top Ethereum transaction network within a month, the Spark large-scale data processing engine is used to count the number of DApp users, the total number of external transactions between accounts, and the number of times smart contracts are used within a month. Statistical distribution histograms are then plotted. Based on the inflection points of each distribution chart, thresholds are determined for the number of DApp users, the amount of external transactions, and the number of times smart contracts are used. Based on these thresholds, external transaction data with transaction amounts exceeding the thresholds, smart contracts with transaction counts exceeding the thresholds, and DApps with more users than the thresholds are filtered out to obtain the top Ethereum transaction network within a month.

3. The method for constructing a 3D visualization based on Ethereum big data according to claim 1, characterized in that, The three-dimensional spatial coordinates of users and tokens of the DAPP and the smart contract points of the DAPP are determined by the spherical layout method and the in-sphere layout method, respectively, and the three-dimensional spatial stable coordinates of all nodes in the Ethereum transaction network are obtained.

4. The method for constructing a 3D visualization based on Ethereum big data according to claim 1, characterized in that, The calculations for attraction, repulsion, and central force are as follows: a. The formula for calculating gravity is: b. The formula for calculating repulsive force is: c. The formula for calculating the central force is: .

5. The method for constructing a 3D visualization based on Ethereum big data according to claim 1, characterized in that, Using an octree data structure, the network model in the initialized 3D space is recursively divided into 8 sub-regions. If a sub-region does not contain any nodes, it is not processed in the subsequent recursion. If a region contains only one node, it becomes a leaf node in the octree. If a region contains two or more nodes, it is further divided using the octree data structure. Finally, all sub-regions construct a non-complete octree. Each node in the tree has at most 8 child nodes, and the leaf nodes in the tree have exactly one node in the Ethereum transaction 3D network.

6. The method for constructing a 3D visualization based on Ethereum big data according to claim 5, characterized in that, Based on the Barnes-Hut algorithm, nearby objects are grouped and treated as a whole. If the distance between objects is large enough, the effect of gravity is approximated by the centroid. The centroid of a group of objects is the average position of all objects in the group, and the mass of the objects is used as the weight. If the positions of two objects are ( x 1, y 1, z 1) and ( x 2, y 2, z 2), the masses are respectively m 1 and m 2. Its total mass and total centroid are respectively: Total mass: Overall centroid: .

7. The method for constructing a 3D visualization based on Ethereum big data according to claim 5, characterized in that, The Barnes-Hut tree is constructed as follows: If the root node x does not contain an object, then place the new node b into it; If the root node x is an internal node, i.e. a non-leaf node, then update the total mass and centroid of x, and recursively insert node b into one of the eight regions; If the root node x is an external node, i.e. a leaf node, containing a node c, then it means that there are two nodes b and c in the same region. The region is further divided into eight sub-regions, and then nodes b and c are recursively inserted into the corresponding branches. Finally, the centroid and total mass of node x are updated. The repulsive forces between nodes are calculated using the constructed Barnes-Hut tree structure. The specific steps are as follows: 1) If the current node is a leaf node and the node is not b, calculate the force it exerts on b and add the result to the net force on b. 2) Otherwise, calculate the ratio s / d. If s / d0 < θ, where s is the width of the region where the internal node is located, d is the distance between the object and the centroid of the node, and θ is the threshold, treat the internal node as a node and calculate the force exerted by the node on b. Add the result to the net force of b. 3) Otherwise, recursively run steps 1) and 2) on the child nodes of this node.

8. A 3D visualization system based on Ethereum big data, characterized in that, It includes an Ethereum transaction data acquisition and storage module, a data cleaning module, a 3D visualization model construction module, and a 3D visualization model display module; The Ethereum transaction data storage module is used to break down the Ethereum transaction big data obtained from the OpenEthereum Ethereum third-party wallet client into different themes and primary keys using EthereumETL tools, and then store it in the HDFS distributed file system under the Hadoop ecosystem for persistent storage. It also uses web crawling technology to obtain DApp data information from multiple websites and stores it in the relational database MySQL for persistent storage. The data cleaning module utilizes the Spark large-scale data processing computing engine to filter out external transaction data with transaction amounts exceeding a threshold, smart contracts with transaction counts exceeding a threshold, and DApps with a number of users exceeding a threshold within a month from the external transaction information table, account balance table, token balance table, and DApp information table in the Ethereum transaction data, thereby obtaining the top Ethereum transaction networks within a month. The 3D visualization model building module is based on the data results processed by the data cleaning module. It adopts an optimized force-directed layout algorithm and uses the octree algorithm for further optimization to build an initial 3D transaction network of full data. Then, a spherical layout method is used for local areas to place the unique users of the DApp on its surface to obtain a 3D Ethereum transaction network model of full data. The optimization process of the force-guided layout algorithm is as follows: First, based on the node attributes in the Ethereum transaction network, nodes are generally divided into three categories: EOA account nodes, smart contract nodes, and token nodes. Smart contract nodes and token nodes are further subdivided according to the type of DApp they belong to: games, finance, social, and exchanges. Based on the relationship attributes in the Ethereum transaction network, edges are generally divided into five categories: external transaction relationships, user-invoiced smart contract relationships, user-held token relationships, smart contract and DApp ownership relationships, and token and DApp ownership relationships. Secondly, the coordinates of all nodes in three-dimensional space are randomly initialized; the nodes are treated as electrons, so that there is a repulsive force between all nodes, and the relationship is treated as a spring, so that there is an attractive force between the nodes in the relationship. Then, the formulas for calculating attraction and repulsion are modified to reduce the repulsion between smart contracts and tokens belonging to the same DApp and increase the attraction between smart contracts and tokens belonging to the same DApp. At the same time, the formula for calculating the central force is modified to make the entire network layout tighten towards the center and to use the Barnes-Hut tree structure to calculate the repulsion between nodes. Under the combined action of the repulsive and attractive forces, the node coordinates are gradually adjusted to construct a three-dimensional network topology. The three-dimensional network topology iterates multiple times under the action of repulsive force, attractive force and central force, and finally reaches an equilibrium state. The coordinates of all nodes in the three-dimensional space are basically fixed, and the stable three-dimensional space coordinates of all nodes are obtained. The coordinates of each DAPP in the three-dimensional space are calculated according to the smart contract and token position of the DAPP. The 3D visualization model display module is based on the Ethereum transaction 3D network model generated by the 3D visualization model building module, and is drawn and rendered using ThreeJS and / or WebGL web-based 3D visualization technology.

9. A computer device, characterized in that, It includes a processor and a memory, the memory being used to store a computer-executable program, the processor reading the computer-executable program from the memory and executing it, and the processor executing the computer-executable program is able to implement the three-dimensional visualization construction method based on Ethereum big data as described in any one of claims 1 to 7.

Images