AI for All
AI for All represents the Mifos strategic initiative and forward-looking roadmap for data science effort around artificial intelligence and machine learning for use cases powered by solutions built on Mifos and Fineract.
In GSOC 2018-20, three use cases that were developed using AI/ML - Credit Scoring, Chatbot and PPI.
Join our bi-weekly AI and ML working group at https://discourse.mifos.org/g/AI-ML to track progress or pick out a task from those described in this section.
You can also view our presentation at the 2020 ApacheCon at:
Use Cases
Use Case Theme | Explanation | Type |
---|---|---|
Cash flow prediction | Predict future/present value of money and company’s inflows/ outflows for projected income/expenses over a period of time | Back office |
Credit scoring | Determine chances a borrower will fail to make a payment on a loan and identity credit risk, cost of debt, interest rate, credit score | Back office |
Cost of equity, risk | Calculate the return potential stockholders expect before investing in a company | Back office |
Portfolio management | Which among a given set of investment decisions are most beneficial for a company by estimating opportunity cost, net present value, internal rate of return, payback period | Back office |
Enterprise value | Calculate a firm's market valuation and liquidation value | Back office |
Spend analysis | Spend management by tracking spending patterns of individuals, with saving suggestions and financial advice for ROI | Back office |
Fraud detection | Audit transactions to identify fraud risk in credit cart application process, monitor suspicious account behavior, false positives like money laundering | Back office |
Security | Analyse large amounts of data to identify risk patterns and improve security | Back office |
Algorithm trading | Tracking market changes and fluctations in stock prices, high frequency trading in hedge funds / investment firms for trade decisions and manage stock volatility | Back office |
Virtual Assistant & Process Automation | Automate manual and repetitive processes, approval/rejection, paperwork with chatbots, call center automation for cost reduction and ops efficiency | Front end |
Product recommendation & Campaign Management | Predict customers with propensity to churn, their RFM score & lifetime value and retarget / win them back with personalised offers, insurance plans, lending etc. | Back office |
Customer Experience based on Customer persona | Track customer behavior and usage across apps / web / social media activity and improve product experience and onboarding flow | Front end |
NPA Management | Detect, measure, predict, and anticipate, among other things, market volatility, liquidity risks, financial stress, housing prices, and unemployment to reduce non-performing assets | Back office |
Lending & Loan Management | Processing, disbursement and fulfilment of loans across borrowers and lenders E.g Home Loan Takeover, Home Loan Top-up, Loan against property | Back office |
Loan Underwriting | Lender determining if a borrower's loan application is an acceptable risk. Underwriters assess the borrower's ability to repay the loan based on an analysis of their credit, capacity, and collateral. | Back office |
Early warning system | Early Warning provides info to financial institutions about consumers' banking activity and history to help them detect fraud and assess risk | Back office |
Working Capital Enhancement | Maximise (Assets - Liabilities) | Back office |
Language Translation | Banks and financial institutions must translate large amounts of business documents from English into Spanish, Chinese, Portuguese, or other languages with quality & speed as well as consistent branding & legality in order to meet international customer expectations. | Front end |
Adaptive Authentication | Adaptive authentication mechanisms evaluate the risks surrounding a particular user login attempt and dynamically steps up the authentication or prevents the login, based on suspicious activity. | Front end |
Social Media Engagement | Automated Responses to Social Media posts and queries by prospects / clients / customers | Front end |
Collateral document management | Scanning & identification of what can be committed as collateral for a transaction to be performed | Back office |
Money Management | AI-driven Reconciliation, EWS, AML, Income leakage, Parking Accounts, Footfall Model, ATM Chargeback, IP infringement | Back office |
Payment channel for making / accepting payments | Based on date, time, nature of transaction | Front end |
References
White Papers on Artificial Intelligence in Banks
Artificial Intelligence in Banks
Redefine Banking with Artificial Intelligence
Artificial Intelligence Applications in Financial Services, Marsh & McLennan Companies
Artificial Intelligence & Financial Services by Apis Partners
Links
https://www.idrbt.ac.in//assets/publications/Best Practices/2020/AI_2020.pdf
https://www.mindtitan.com/case/artificial-intelligence-in-finance-and-banking/
https://www.paymentsjournal.com/the-18-top-use-cases-of-artificial-intelligence-in-banks/
https://www.fintechnews.org/ai-and-machine-learning-in-finance-use-cases-in-banking-insurance-investment-and-cx/
https://builtin.com/artificial-intelligence/ai-finance-banking-applications-companies
https://marutitech.com/ai-and-ml-in-finance/
https://towardsdatascience.com/ai-use-cases-for-finance-and-banking-716b43082c5b
https://www.wipro.com/en-IN/business-process/why-banks-need-artificial-intelligence/
https://www.datasciencecentral.com/profiles/blogs/19-ai-amp-iot-use-cases-in-banking-industry
https://www.mckinsey.com/industries/financial-services/our-insights/ai-bank-of-the-future-can-banks-meet-the-ai-challenge#