Application of Data science in Banking sector

Transcript

1. DATA SCIENCE: BANKING SECTOR ANALYZING THE ROLE OF DATA SCIENCE IN BANKING SECTOR PRANOY KRISHNA
2. THE BANKING SECTOR • Banking sector is one of the most important sectors in a growing economy like India. It has been instrumental in the level of digitalization India has achieved throughout the last few decades. • One of the major propellants to this growth in India was certainly the incorporation of modern processes like Data analysis. • One of the major issues that the banking sector has been facing for the past few decades was the issue of NPAs. Through the advent of data science and data analysis, banks have been successful in reducing their NPAs and making sure that the right person is receiving the credit.
3. MAJOR APPLICATIONS OF DATA SCIENCE IN THE BANKING SECTOR
4. RISK ANALYSIS • One of the major roles of a bank is to act as an efficient investment advisor. They maximize the effectiveness in this field by utilizing data science and has labelled it as “Risk Analysis”. It helps us to Identify and Assess risks and also to develop adequate preventive measures. • Through this analysis they are aiming to differentiate between risky and non risky assets in a better way. This would help in making adequate advices by putting the risk appetite of the investor into consideration. • It effectively automizes the data analysis for having a reduction in losses. It has been successful in reducing the amount of NPAs that are existing in the banking system. What it does basically is to understand and predict the flow of transactions and provide themselves with the necessary information needed to understand whether the following investment or credit has been “Risky” or “Non risky”
5. MARKETING & SALES PERFORMANCE ANALYSIS • Marketing is also an important pillar of a successful banking structure. They should be successful in determining what all services that a customer is likely to avail on the basis of his or her interests. • It can help the bank to analyze and find out who all are their most profitable customers through which they can grant personalized services for increasing profit. • It can help in pinpointing what all areas the bank has fallen short of in a given year. For example if some of their services are not availed in a given year they can do a marketing analysis to find out the potential customers they can have with respect to that service. • Sales performance analysis is another area in which data analysis can be effectively utilised. If a particular investment or a purchase is made , it is very important to analyse how impactful this has been to the institution from a profitability standpoint of view. It can be further extended to analyze how many purchases of this sort has to be made for the bank to reach at a desired level of performance.
6. CUSTOMER SUPPORT • Data analysis is immensely helpful in this particular field. Customer grievances can be better understood and adequate actions can be undertaken. For example, if there are lot of queries towards a particular service offered by the bank, the bank can come up with an alternative way to simplify the process, hence making it more user friendly. • Speedy resolution of user problems are another added advantage of incorporating data science in this particular area. Customers no longer have to wait for a long time as facilities like Chat Bots have made the process a whole lot more faster.
7. FRAUD DETECTION • Through this method the banks can effectively analyze the unusual changes in the behavior of transactions and can quickly report it, and investigate it. This would be helpful in tackling the problem at a very early stage. Even the biggest of the biggest scams can be easily averted once it is detected at an early stage. • Accurate reports are generated on the basis of different statistical parameters such as standard deviations and averages. Unusually low and high numbers are noted. • Methods such as duplicate testing, Gap testing, Data validation are all utilized to produce accurate results.
8. CASE STUDIES
9. FRAUD ANALYSIS IN THE BANKING SECTOR: Danske Bank. • Danske bank is the largest bank in the country of Denmark. It was struggling with a low 40% fraud detection rates and a 99.5% rate of false positives(1200 per day). The management then decided to introduce an analytical model. They integrated the deep learning software with a graphic processing unit (GPU). They partnered with Terdata Consulting, and was successful in attaining a 50% increase in the fraud detection rate and a 60% decrease in the false positive rate. Other than that they saw a huge 70 million dollar profit in the year 2018, right after the change were made. • It is clearly evident from this case study that the usage of data science in this particular bank has resulted in a significant positive change for them. This Photo by Unknown Author is licensed under CC BY-SA
10. JP Morgan Chase – Big Data Analytics • JP Morgan Chase is one of the largest banks in the world. It has a customer base of around 3 billion. They have incorporated a big data framework called Hadoop to leverage the large amount of customer data that they have. They utilize frameworks like these for a large amount of areas like investment projects, risk managements, fraud detections etc. through this they are attempting to combine transaction data of approximately 30 million customers. By the help of these big data analytic frameworks, they are becoming successful in helping their customers to enhance the value of their investments, making them stand apart from their peers. Reports are generated on the basis of the data collected and the clients are made aware of the customer trends that are made out of it.