Data_science_in_Banking.pdf

Transcript

1. Application of Data science in Banking MANIKANTA 2237012
2. Introduction  As we all know Banking is a huge service sector which connects whole lot of people and does services such as accepting deposits and helps people to save and lends money to borrowers  It’s very important that for Bank to be safe , profitable and having liquidity at proper time to fulfill the demand of the customers  It’s very important for a bank to adopt data science methodology which would help banks function smoothly and quickly where the banks would be dealing with millions of customers personal data , their credit data , deposit data etc  There are various problems with solutions explained in further slides
3. Problems and Applications of Data science 1. Fraud detection and prevention Every year financial institutions are spending billions against fraud detection applications, as it may hurt the company’s brand and reputation. Data science plays a key role in collecting, summarizing, and predicting the customer database to detect fraudulent activities. Analysis of customer records to drive accurate information is not possible before the existence of data science/big data. AI and Machine learning can help banks combat fraudulent activities. For example, data models can be built for analyzing credit card frauds that provide data intelligence and classify legit or fraudulent transactions, based on details like purchase amount, location, merchant, time and other parameters.
4. 2. Risk Management  Risk management in banks has changed substantially in the last decade as new threats emerge.  The regulations have also gone stricter post-global financial crisis.  The adoption of data science is enabling new risk, management models.  Machine learning technologies can identify complex, nonlinear patterns in large volumes of data and help create models with higher accuracy. These data models also self-learn with every bit of every data and pattern to improve their predictive power with time.
5. 3.Customer Data Analysis  Banks are collecting large amounts of data from consumers.  Analyzing these datasets is possible with data science technologies.  Based on the information collected through social media, customer surveys and data from other touchpoints, the banks can understand customer sentiment NLP.  Machine learning and data science can deconstruct these data sets easily and provide deep data intelligence on customers’ needs, wants and perceptions about the bank.
6. 4.Marketing & Sales  Data science in banking can help create a personalized window for every customer, by dividing the data into demographical, geographical and historical data sets.  These datasets provide deeper insights regarding how a customer responds to an offer/promotion.  Therefore banks can make personalized outreach to interact with customers.  Machine learning helps in creating powerful recommendation engines that can create upsell/cross-sell opportunities for banks.  The key to success in marketing is to customize an offer that suits particular customer preferences and need
7. 5. AI-driven chatbots & Virtual Assistants  A chatbot is a customer program designed to mimic human conversations on a web. The usage of chatbots in banking decreased the customer waiting time and increased the rate of interaction per minute.  The rule-based chatbot operates on a particular command and an AI-based chatbot gets smarter with every interaction.  Erica ( virtual Assistant Robot) can currently operate on voice and text commands, it can schedule the payments and helps in getting information regarding past transactions.  Erica leading customers for better financial health. Capital one has a similar chatbot that helps customers to manage money via text. This is helping banks save a lot of money. According to Juniper Research, chatbots will help banks save S7.3 billion globally by 2023