Micro-loans have become increasingly popular amongst individuals and SMEs. These customers demand a loan product that is more flexible: much lower loan limits, near-instant loan decision, and more frequent withdraws. While many banks and fintech companies work hard to develop micro-loan products for new customers, their existing credit risk capabilities are unable to meet the new requirements: quickly roll out new risk models to support the new micro-loan product development.
Alibaba Cloud Credit Risk Management Solution allows lenders to easily apply machine learning and use their own dataset to develop, train, and deploy credit risk models. It comes with a scorecard template including workflow, algorithm, and test dataset that makes the variable transformation process highly efficient. Banks and fintech companies can use this solution to proactively manage credit risk, support new micro-loan products, and confidently meet regulatory requirement, such as Basel II.
Challenges and Solutions
For banks or fintech lenders alike, credit risk management has long been challenging in many aspects
Inefficient new risk model build
Credit score model needs to be updated quickly to support business growth. This is challenge due to constraints of tool: it is hard for the modeler to quickly identify the relevant data attribute, model validating needs a lot of manual work including data import, and result analysis; model deploying is time consuming because of manual process.
Real time scoring is needed
Micro-loan user expects to know their limits instantly after application. This requires the lender to invest in server and other compute resources and handle peak concurrent loan applications. However, it can be hard to predict the number of users, when the peak load would take place, and how much hardware to invest.
Inefficient reporting process
Risk management team needs to prepare various credit risk reports that are updated in a timely manner: acquire data from various systems, perform data consolidation, calculate the risk value at borrower level, aggregating it to portfolio level. Traditionally, this is done manually with spreadsheet-based reporting, which often overburdens analysts and IT.
Machine Learning for Model
Using machine learning and other AI technology will help financial organization to quickly identify the attributes which are relevant for credit risk, especially from alternative data where past experience is limited. The solution also comes with scorecard templates and embedded algorithms which are commonly used for credit risk evaluation.
Better On-going Risk Management
Using machine learning and AI saves lots of time of finding the correlation between a data attribute and default. This would speed up the upgrading process. Instead of reviewing the model once a year, it could be done more frequently to ensure the credit model can always capture the changes of the user behavior, help to make the best decisions for loan approval.
Report Credit Risks Efficiently
Cloud computing can process large volumes of data quickly and cost-effectively. This can help financial institutions to calculate risk exposure faster and more frequently at lower cost. By doing this, CROs can better communicate risks exposure to executives, help the organization to make better decisions, and meet regulatory requirement more confidently.
How It Works
Having the right model is one of the most challenging tasks for credit risk management. How can I find the right tool and platform to help overcome those challenges.
Alibaba Cloud Dataworks helps to consolidate data from different resource to a centralized platform.
Platform for AI, which is also used by Ant Financial for its credit risk models and other fine-tuned models, can help modelers to build which are most suitable for the organization.
Model report is available to clearly document the model structure which is required by regulator. Once deployed, the credit risk score calculation would be real time with MaxCompute, even at peak business demands when a large number of customers are applying in parallel.
Platform for AI
An end-to-end platform provides various machine learning algorithms.View Detail
ApsaraDB RDS for MySQL
An on-demand database hosting service for MySQL.View Detail
Legacy technologies and warehouse investments are no longer suitable for increasing demands for risk computation speed, frequency, and aggregation capabilities.
An efficient risk aggregation infrastructure, built on Alibaba Cloud with DataWorks and Maxcompute, enables multiple requests for risk reports in multiple formats to be delivered quickly and smoothly.
The solution provides organization advanced report design and generate tools to meet complex credit risk reporting requirements.
A powerful and accessible data visualization tool.View Detail
Elastic Compute Service (ECS)
An online computing service offers elastic and secure virtual cloud servers.View Detail
LGMS PCI DSS Professional Services on Alibaba Cloud
PCI DSS professional services provider currently offers an assisted, user friendly and easy to use SAQ system on Alibaba Cloud.Learn more >
Credit card bill statements-based-credit scorecard
Scorecard is not only a machine learning algorithm, but also a generic modeling framework used to build a model for assessing credit risks.Learn more >
Security & Compliance Center
We are committed to providing stable, reliable, secure, and compliant cloud computing products and services.Learn more >