Smart Contract Classification Using Visual Transaction Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current block explorers present technical data in textual and/or table forms, which are not user-friendly for users, especially when it comes to understanding smart contracts, as they confuse new users and do not aid in understanding blockchain concepts.
Innovation Solution
Implementing machine learning techniques to classify smart contracts recorded on a transaction log, generating graphical representations of transactions using shapes and lines to distinguish between different types of smart contracts, thereby making the underlying information more understandable.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If technical data is presented in textual and/or table forms, then data completeness is maintained, but user-friendliness deteriorates
Solution Approach 1:
The patent creates visual copies of transaction data through graphical representations. Instead of displaying raw textual data, the system generates visual copies using shapes, colors, and spatial arrangements that represent different aspects of the transaction data, making it more accessible while preserving the underlying information through structured visual encoding.
Solution Approach 2:
The patent transforms one-dimensional textual data into two-dimensional graphical representations. By adding spatial dimensions (position, size, shape, color) to the data presentation, the system enhances user comprehension without losing information, as multiple data attributes are simultaneously represented across different visual dimensions.
2Loss of information
If smart contract code is displayed directly, then information accuracy is maintained, but understanding difficulty increases
Solution Approach 1:
The patent introduces visual elements as intermediaries between the raw smart contract code and the user. Shapes, colors, and graphical representations serve as mediators that translate complex code structures into comprehensible visual forms, allowing users to understand contract logic without directly interpreting programming syntax.
Solution Approach 2:
The patent changes the representation parameters of smart contract data from textual format to visual format. By transforming code into graphical elements with varying properties (shape types, colors, sizes, positions), the system maintains information accuracy while dramatically improving understandability through human-friendly visual parameters.
3Measurement precision
If detailed transaction data is presented, then measurement precision is maintained, but device complexity increases
Solution Approach 1:
The patent segments detailed transaction data into distinct visual components represented by different shapes and colors. Each shape type corresponds to specific contract categories, allowing complex data to be divided into manageable visual segments that can be processed and understood separately, reducing the perceived system complexity.
Solution Approach 2:
The patent uses dimensional transformation to handle data complexity. By representing detailed transaction attributes in multiple visual dimensions (spatial position, shape type, color, size) rather than through nested textual structures, the system maintains measurement precision while reducing the complexity burden on the display and processing system.
Data Source
AI summary
Systems and methods to train and/or utilize a machine learning model to classify smart contracts in transactions recorded on a transaction log stored in immutable distributed electronic storage, are described herein. Exemplary implementations may: obtain transaction information characterizing transactions recorded on a transaction log stored in immutable distributed electronic storage; obtain classification information identifying contract classes of the transactions; aggregate transaction information and classification information into model training information; provide the model training information to a machine learning model to train the machine learning model and generate a trained machine learning model; and/or perform other operations. Exemplary implementations may: provide transaction information of new transactions as input into a trained machine learning model; obtain output from the trained machine learning model; generate, from the output, classification information including contract classes of the new transactions; and/or perform other operations.