# 信贷风控 > Design an end-to-end machine learning system for a real-time loan approval/rejection model, such as credit cards. Discuss the infrastructure, features, model, training and evaluation aspects of the system. ## 1. requirements **functional** * stage: 获客、贷前(Loan origination)、贷中(Loan maintenance /servicing)、贷后(Delinquency management/recovery) * types of loans it will support * types of risk it will support -> 决定了是一个什么样的机器学习任务, 影响难点 * 目标: 识别优劣(风险，需求，价值) * 业务指标 **non-functional** * compliance requirements * scalability goals * reliability, security ## 2. ML task & pipeline & keys 信贷风控决策流 黑白名单 rule + model binary classification, multi class classification or multi label classification ## 3. data * user (关系型数据库) * credit 信用 * fraud 欺诈 * log (分布式文件系统) * label * 滚动率(Roll Rate)、vintage * 从数据指标中发现风险点 * 信贷: 逾期不放贷款的就是黑样本 需求类: 履约风险: 履约能力: 外部数据 * 因合规要求，很多互联网大数据只能从征信平台以评分的形式引入 ## 4. feature > feature engineering is the art > **user** * ID/Address Proof: Voter ID, Aadhaar, PAN Card * Employment Information, including salary slips * Credit Score * Bank Statements and Previous Loan Statements * 反欺诈重要feature，如ip，device\_id，idfv，phone number *** * graph * 行为序列 * compliance -> buy some user feature ![](https://2388792444-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FELc1F32q81Oj7PEFYHDV%2Fuploads%2Fgit-blob-5c12ddc11b8231b63b3f435cf2b8ecbf8499f7be%2F03ml-loan-feature2.png?alt=media) **稳定性** * SHAP 是一种解释模型输出的方法，能展示某个变量对预测值的贡献。检查：变量的值是否与模型输出呈合理的趋势；是否存在“非预期拐点”或“跳跃”（可能是过拟合的信号）。 * 某些变量和风险标签的关系应该是单调的（如负债率↑，违约概率↑），否则可能存在逻辑问题、数据问题，或模型过拟合。建模阶段通过分析变量与 label 的WOE 图（Weight of Evidence）或 KS 图判断是否具有单调关系 ## 5.model > 可解释性和稳定性（stability）。模型不仅要在当前样本上表现好，还要对未来数据、不同时间段、不同人群具有鲁棒性，也就是稳定性。 ![](https://2388792444-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FELc1F32q81Oj7PEFYHDV%2Fuploads%2Fgit-blob-5a4d2390c192e39d59a5ed9ed045114673dcc95a%2F03ml-loan-train.png?alt=media) * Credit Scoring Models 评分卡 * LR * GBDT * NN * Probabilistic Calibration ## 6. evaluation * offline * 准确率、AUC、Log Loss、Precision、Recall * Kolmogorov-Smirnov，风控常用指标 ## 7. deploy & serving ![](https://2388792444-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FELc1F32q81Oj7PEFYHDV%2Fuploads%2Fgit-blob-8c6ebe63c7fafda5e3c393ad36a3d82e0eca1027%2F03ml-loan-inference.png?alt=media) * feature service * prediction service ## 8. monitoring & maintenance * Approval Rate ## 9. QA & optimization * cold start * profit/revenue come from which (credit score) part of customers, risk come from which part of customers ## reference * [外部数据在信贷风控和经营中的应用](https://mp.weixin.qq.com/s/bJGFsIuNVNfg1866PQ9gYg) * [Real Time Risk Management and Assessment](https://www.gigaspaces.com/blog/real-time-risk-management) * [How Loan Management Software Improves Loan Risk Assessment For Lenders](https://lendfusion.com/blog/loan-risk-assessment/) * [我现在来演示一下风控模型的架构 - 风控少侠的文章 - 知乎](https://zhuanlan.zhihu.com/p/717674909) * [风控架构简单分享 - 回首乱山横的文章 - 知乎](https://zhuanlan.zhihu.com/p/683754431) * [特征工程到底是什么？](https://www.zhihu.com/question/29316149) * [风控模型—区分度评估指标(KS)深入理解应用 - 求是汪在路上的文章 - 知乎](https://zhuanlan.zhihu.com/p/79934510) * [小样本学习在滴滴治理和安全场景应用](https://mp.weixin.qq.com/s/sf5B0hmhwIoYLuCU7Rdvug) * [复杂风控场景下，如何打造一款高效的规则引擎](https://tech.meituan.com/2020/05/14/meituan-security-zeus.html) * [Build and Deploy a Real Time ML System](https://paulabartabajo.substack.com/p/implementing-a-real-time-ml-system) * [Let's build another Real Time ML System](https://www.realworldml.net/blog/let-s-build-a-real-time-ml-system-2) *