区块链智能合约solidity数据集

该数据集包含7种常见漏洞类型（重入、时间戳依赖、整数溢出、tod、tx.origin、unchecked-send、unhandled-exceptions和安全合约）的区块链智能合约 Solidity 源代码样本，针对稀缺漏洞类型，我们使用了基于模版的数据样本增强技术以平衡数据分布，适用于神经网络模型的训练、验证与测试，旨在支持智能合约漏洞检测、分类及安全分析研究。

This dataset contains Solidity source code samples of blockchain smart contracts covering 7 common vulnerability types: Reentrancy, Timestamp Dependency, Integer Overflow, tod, tx.origin, unchecked-send, unhandled-exceptions, as well as secure contracts. To balance the data distribution, template-based data sample augmentation techniques were adopted for scarce vulnerability types. This dataset is applicable for the training, validation, and testing of neural network models, and aims to support research on smart contract vulnerability detection, classification, and security analysis.