通信用户通话特征数据集

本数据集包括用户号码、日/月通话汇总（如通话次数、平均时长、通话离散度等），旨在精准识别潜在的诈骗行为，并即时向可能受害的群众发出预警，有效保障其财产安全。 我们确保所有处理过程均遵循严格的隐私保护标准，仅提取必要的数据特征用于分析，且处理后的数据以匿名形式返回，仅用于评估用户的近期通话模式，以便快速响应任何可疑诈骗活动。 通过深度挖掘这些通话特征，结合大数据分析与模型预测，我们能够更精准地识别诈骗行为模式，同时提升对潜在受害者的预警效率。随着系统的持续运行，数据集合将不断迭代更新，为模型提供更加丰富和准确的训练样本，从而进一步优化诈骗行为的识别与预防能力，实现风险的提前预判与有效遏制。

This dataset includes user phone numbers, daily and monthly call summaries (such as call count, average call duration, call dispersion, etc.). It aims to accurately identify potential fraudulent activities, send real-time alerts to potentially victimized individuals, and effectively protect their property safety. We ensure that all processing procedures comply with strict privacy protection standards, only necessary data features are extracted for analysis, and the processed data is returned in anonymous form, which is exclusively used to evaluate users' recent call patterns to quickly respond to any suspected fraudulent activities. By deeply mining these call features and combining big data analysis and model prediction, we can more accurately identify the patterns of fraudulent behaviors, while improving the efficiency of alerting potential victims. As the system operates continuously, the dataset will be iteratively updated, providing richer and more accurate training samples for the model, thereby further optimizing the ability to identify and prevent fraudulent behaviors, and achieving early prediction and effective containment of risks.

搜集汇总

数据集介绍

特点

通信用户通话特征数据集由中国电信股份有限公司山东分公司提供，主要用于反欺诈模型的构建。该数据集包含用户号码、日/月通话汇总等信息，旨在通过分析通话特征识别和预防诈骗行为，同时确保用户隐私得到保护。