Impact Of Big Data on The Banking and Finance

Sector
Akriti Sharma
BBA (Banking and Insurance), Shoolini University

Abstract— In the current era of information technology, Keywords— Big Data, Banking, Risk Management,
big data has become increasingly important in all
Fraud Detection, Data Usability, Customer Segmentation
industries around the world. Big Data is known for its
and targeting
ability to facilitate effective decision making. Companies
all over the world are now using Big Data for market and I. INTRODUCTION
customer analysis; however, big data usage in the
financial sector is much higher. The banking sector has Data science has had a significant impact on the
been impacted by the emergence of big data. With the banking and financial sector. Advanced analytics is
help of big data analytics, banks can gain insights to enabling superior performance in the organizations
customer behaviour, detect fraud, and make more willing to make the proper commitment. Across all
informed decisions. Additionally, big data has also industries, companies that are more analytically driven
enabled banks to personalize their services and products realize financial growth three times higher than their
based on individual customer needs and preferences. This less analytical competitors. The banking sector, with a
has resulted in increased customer satisfaction and long history in leveraging data, starts from the strongest
loyalty, leading to higher profits for the banks. Every position. Some of the key innovations driven by data
minute, banking and financial institutions create and science and AI in Finance include open banking,
trade massive amounts of money all around the world. As
blockchain, mobile banking, and AI and machine
a result, this industry is data intensive. The data in this
sector is complex information that falls within the learning. These technologies have improved and
purview of Big Data, which has already been automated the way third parties draw inferences from
characterised as "a large, diverse, and complex set of data and thus how financial institutions make data-
information that is growing at an increasing rate." These driven decisions.
data can help banks gain a better knowledge of their As the internet, smart phones, and other apps have
client base, product performance, and industry trends in grown in popularity,
this competitive market area. The volume of digital data has increased. The
With the growth of technology over the last decade, half enormous potential of using this knowledge, also known
of the adult population has begun to use digital banking. as Big Data, is acknowledged by Private companies and
The ever-increasing number of tablets, smartphones, and governments alike must work together to provide
other electronic gadgets has made it simple. Considering
genuine value for customers while increasing
the importance of big data in financial sector, this paper
is an attempt to conduct a comprehensive literature productivity over time. Big data may transform
review in the field of big data and finance. As a result, the businesses and economies, but data science is the true
study will add to the body of knowledge by opening new game changer.
avenues for empirical research in the fields of big data Big Data Analytics is the study of large and diverse
and finance. data sets to identify hidden patterns, consumer wants
and trends, unknown correlations, customer wishes, and
other useful knowledge that aids in the achievement of
significant marketing goals. The astonishing increase is
due to both the number of data generating devices and
the number of sensors in each device; around 11 billion  Unstructured Data
devices.  Semi – Structured Data
Demonetization ushered in a new age of tremendous
growth in India's e-banking and e-commerce. Following
demonetization, the Indian government began to A. Structured Data
encourage digitalization to create a cashless society, Structured data refers to data that is organized and
paving the path for increasing eBanking and e- formatted in a specific way to make it easily readable
Commerce transactions. There are just a few mobile and understandable by both humans and machines. This
payment apps, such as Mobi Kwik, Paytm, Freecharge, is typically achieved using a well-defined schema or
Google Pay, BHIM UPI, State Bank Buddy, ICICI data model, which provides a structure for the data.
Pockets, HDFC Chillr, Citi MasterPass, Vodafone M- Structured data is usually stored in well-defined
pesa, and others. The business's potential may be readily schemas such as databases. It is tabular with columns
approximated by looking at Paytm’s growth - an e- and rows that clearly define its attributes. Each data
Commerce website launched in 2010, offering solely element is typically assigned a specific field or column
mobile recharging, managed to gain personal in the schema, and each record or row represents a
investment from billionaire Ratan Tata by March 2015. specific instance of that data. It is highly valuable
The same month, Alibaba Group of China invested because it can be easily searched, queried, and analysed
$575 million in the firm. Paytm became India's largest using various tools and techniques. This makes it an
mobile payment service platform in November 2016, ideal format for data-driven applications such as
with over 150 million wallets and 75 million Android business intelligence and analytics, as well as for
app installs. machine learning and artificial intelligence applications.
B. Unstructured Data
II. WHAT IS BIG DATA? Unstructured data refers to information that does not
Big Data is such a vast and complicated collection of have a pre-defined data model or is not organized in a
data sets that it is hard to manage using one-handed pre-defined manner. It is typically text -heavy but may
information management tools or traditional data contain data such as dates, number, and facts as well.
processing software. Data was historically handled on Since unstructured data does not have a predefined
servers since the volume of data was modest and data model, it is best managed in non-relational
processing time was limited. However, in today’s (NoSQL) databases. Another way to manage
modern technology world, data moves at breakneck unstructured data is to use data lakes to preserve it in
speed, and people are increasingly reliant on it. At the raw form.
rate at which data is growing, it is becoming difficult to
store it on any server. Data warehousing has C. Semi-structured Data
traditionally been used to store vast amounts of data and
Semi-structured data is a type of data that is not
perform complicated analytics. The data centre, on the
purely structured, but also not completely unstructured.
other hand, works on a limited collection of data, which
It contains some level of organization or structure but
leads to inaccuracy. If you modify the sample, you
does not conform to a rigid schema or data model and
could obtain the same result. It takes a long time to
may contain elements that are not easily categorized or
obtain substantial results. A data centre does not store
classified.
and process massive amounts of data in a brief period.
Semi-structured data is typically characterized using
These contribute to the advancement of Big Data and
metadata or tags that provide additional information
Big Data Analytics. The difficulty of big data is to
about the data elements. For example, an XML
capture, store, search, share, transmit, analyse, and
document might contain tags that indicate the structure
visualise this data.
of the document but may also contain additional tags
that provide metadata about the content, such as author,
Following are the type of Big Data:
date, or keywords. It is becoming increasingly common
 Structured data
as organizations collect and process more data from a The banking sector in India has a long and complex
variety of sources, including social media, IoT devices, history that can be traced back to the ancient times.
and other unstructured sources. While semi-structured According to some sources, the concept of banking
data can be more challenging to work with than strictly originated in India around 2000 BCE, when merchants
structured data, it offers greater flexibility and and moneylenders used to lend money to farmers and
adaptability, making it valuable tool for data analysis traders. The earliest forms of banking were informal
and management. and based on trust and social relations.

The formal banking system in India emerged during

III. CHARACTERISTIC OF BIG DATA the colonial period, when the British established the
first banks in the country. The Bank of Hindustan was
Big data is a phenomenon that has no perfect definition,
the first bank to be set up in India in 1770, followed by
but three dimensions have been identified that are
the General Bank of India in 1786 and the Bank of
recognised by all definitions. these three dimensions are
Bengal in 1806. These banks were involved in financing
data volume, data velocity, and data variety. the
trade and commerce between India and Britain.
massive volume of data is generated at an accelerated
rate, known as increasing velocity, and is gathered
The banking sector in India underwent a major
through various sources. the term "big data" is defined
transformation after independence in 1947, when the
as a data set that is beyond the ability of typical
government initiated various reforms and regulations to
database software tools to capture, store, manage, and
promote economic development and social welfare. The
analyse. the 5 v's of big data represent the five different
most significant reform was the nationalization of 14
dimensions: volume (volume), velocity (speed), variety
major commercial banks in 1969, which aimed to
(volteness, scalability, and variety) and value.
increase the outreach and accessibility of banking
1) Volume: Big data has a massive amount of data services to the rural and poor sections of the society.
that can range from terabytes to zettabytes. The volume The nationalized banks were also entrusted with the
of big data poses challenges for storage, processing, and responsibility of implementing various priority sector
analysis lending schemes, such as agriculture, small-scale
2) Velocity: Big data is generated at a high speed industries, education, housing, etc.
and needs to be processed and analysed in real time or
near real time. The velocity of big data poses challenges The banking sector in India has evolved over the
for streaming, ingestion, and latency. years to meet the changing needs and demands of the
economy and the society. The liberalization and
3) Variety: Big data comes from various sources and globalization policies of the 1990s opened new
formats, such as structured, unstructured, semi- opportunities and challenges for the banking sector, as it
structured, text, image, video, audio, sensor, etc. The had to compete with foreign and private banks that
variety of big data poses challenges for integration, entered the Indian market. The banking sector also
transformation, and quality. witnessed rapid technological innovations and
4) Veracity: Big data has various levels of reliability, digitalization, such as internet banking, mobile banking,
accuracy, and trustworthiness. The veracity of big data ATMs, etc., that enhanced the efficiency and
poses challenges for validation, cleaning, and security. convenience of banking services.
5) Value: Big data has the potential to provide Today, the banking sector in India is one of the
valuable insights and benefits for various domains and largest and most diversified in the world, comprising of
applications. The value of big data poses challenges for public sector banks, private sector banks, foreign banks,
extraction, visualization, and monetization. regional rural banks, cooperative banks, microfinance
IV. BANKING SECTOR IN INDIA institutions, non-banking financial companies, etc. The
banking sector plays a vital role in supporting the
economic growth and development of India by
providing credit and financial services to various sectors IoT, human interaction information and position
and segments of the society. information in the Internet world.
Innovative financial practises have also increased in The pace of data generation continues to accelerate.
the new century. Banks have also benefited from the Much of this data it is not generated from the
technology revolution that popularised computers and business transactions that happen in databases-
mobile phones and brought the Internet to the public. instead, it comes from other sources, including cloud
Beginning with internet banking, mobile banking, and systems, web applications, video streaming and
remote banking, technology has enabled banks to smart devices such as smartphones and voice
reduce their reliance on physical branches while assistants. This data is unstructured and in the past
reaching out to a larger consumer base through virtual was left mostly unprocessed and unused by
banking services. organizations.

VII. APPLICATION OF BIG DATA IN BANKING SECTOR

V. ADVENT OF BIG DATA Big data has many applications in the banking sector.
Big data has revolutionized the modern business Some of these include:
environment in recent years. A mixture of structured,  Risk management
semi-structured and unstructured data, big data is a  Fraud detection
collection of information that organizations can mine  Customer segmentation
for business purposes through machine learning,  Sentiment analytics
predictive modelling, and other advanced data analytics  Examine customer feedback
applications. 1) Risk management: Risk management in banking
Big data analytics builds on concepts that have been sector using big data. Risk management is a
around for centuries. The history of data analysis that crucial process for banks to ensure their financial
led to today’s advanced big data analytics starts back in stability and profitability. Banks face distinct
the 17th century in London. In 1663, John Graunt types of risks, such as credit risk, market risk,
introduced statistical analysis with the bubonic plague. operational risk, liquidity risk, and reputational
The London haberdasher published the first collection risk. To manage these risks effectively, banks
of public health records when he recorded death rates need to collect, analyze, and monitor enormous
and their variations during the bubonic plague in amounts of data from various sources, such as
England. customer transactions, market movements,
The evolution of modern technology is interwoven regulatory reports, and social media. Big data is a
with the evolution of Big Data. Such foundational steps term that refers to the massive volume, variety,
to the modern conception of Big Data involve the velocity, and veracity of data that is generated and
development of computers, smart phones, the internet, processed by modern technologies. Big data can
and sensory (Internet of Things) equipment to provide help banks improve their risk management by
data. Credit cards also played a role, by providing providing them with more accurate, timely, and
increasingly copious amounts of data, and certainly comprehensive information about their exposures,
social media changed the nature of data volumes in performance, and opportunities. Big data can also
novel and still developing ways. enable banks to use advanced analytics
VI. GENERATION OF BIG DATA techniques, such as machine learning, artificial
intelligence, and natural language processing, to
Big data is generated through longitudinal and
extract valuable insights and predictions from the
distributed data sources. Such data sources include
data.
sensors, videos, click streams, and/or all other
available data sources. At present, main sources of
Some of the benefits of using big data for risk
big data are the operation and trading information in
management in banking sector are:
enterprises, logistic and sensing information in the
 - Enhancing credit risk assessment and scoring by Banks collect and store massive amounts of data
using alternative data sources, such as social from various sources, such as customer profiles,
media, mobile phone usage, and online behaviour, transaction records, online behaviour, social
to complement traditional credit bureau data and media, geolocation, and more. This data is often
improve customer segmentation and default unstructured, noisy, and heterogeneous, making it
prediction. difficult to analyze and extract meaningful
- Reducing market risk by using real-time data insights. Moreover, fraud patterns are constantly
feeds and sentiment analysis to monitor market evolving and adapting to avoid detection,
trends and volatility and adjust portfolio requiring banks to update their fraud detection
allocation and hedging strategies accordingly. models frequently.
- Mitigating operational risk by using anomaly Big data technologies can help banks overcome
detection and fraud prevention algorithms to these challenges and enhance their fraud detection
identify and prevent fraudulent transactions, capabilities. Big data technologies refer to a set of
cyberattacks, and human errors. tools and methods that can process and analyze
- Optimizing liquidity risk by using cash flow large and complex data sets efficiently and
forecasting and scenario analysis models to effectively. Some of the benefits of using big data
anticipate and manage cash inflows and outflows technologies for fraud detection are:
under different market conditions and regulatory - Scalability: Big data technologies can handle
requirements. massive volumes of data and scale up or down as
- Managing reputational risk by using sentiment needed.
analysis and natural language processing to - Speed: Big data technologies can process data in
monitor customer feedback and public opinion on real time or near real time, enabling banks to
social media platforms and online forums and detect and respond to fraud quickly.
respond proactively to any negative comments or - Variety: Big data technologies can handle
complaints. several types of data, such as structured,
unstructured, semi-structured, text, image, video,
Big data is transforming the way banks manage audio, etc., and integrate them into a unified view.
their risks and enhance their competitive - Accuracy: Big data technologies can apply
advantage in the dynamic and complex financial advanced analytics techniques, such as machine
environment. However, big data also poses some learning, deep learning, natural language
challenges and risks for banks, such as data processing, etc., to discover hidden patterns and
quality, security, privacy, governance, ethics, and anomalies in the data and improve the accuracy of
regulation. Therefore, banks need to adopt a fraud detection models.
comprehensive approach to leverage big data for - Innovation: Big data technologies can enable
risk management while ensuring compliance with banks to explore new sources of data and new
ethical standards and legal frameworks. ways of detecting fraud, such as sentiment
2) Fraud Detection: Fraud is a severe problem for the analysis, behavioural biometrics, social network
banking sector, as it can cause financial losses, analysis, etc.
reputational damage, and legal consequences.
Fraudsters use various techniques to deceive In conclusion, fraud detection in banking sector
banks and their customers, such as identity theft, using big data is a promising approach that can
phishing, malware, card skimming, money help banks reduce fraud losses, protect their
laundering, and more. To combat fraud, banks customers, and comply with regulations.
need to employ effective fraud detection systems However, it also poses some challenges and risks
that can identify and prevent fraudulent that need to be addressed carefully, such as data
transactions in real time. quality, security, privacy, ethics, governance, etc.
One of the challenges of fraud detection is the
large and complex data that banks must deal with.
3) Customer segmentation: Customer segmentation - Improve customer satisfaction and advocacy by
is a process of dividing customers into groups listening to customer voices and opinions and
based on their characteristics, behaviours, and taking actions to improve customer experiences.
preferences. It helps banks to tailor their products, - Optimize marketing strategies and campaigns by
services, and marketing strategies to different segmenting customers based on their sentiments
segments and increase customer satisfaction and and preferences and delivering targeted and
loyalty. Big data is a term that refers to large and customized messages that resonate with them.
complex datasets that are generated from various - Mitigate risks and frauds by monitoring
sources, such as transactions, social media, customer sentiments and detecting anomalies or
sensors, and mobile devices. Big data can provide suspicious activities that indicate fraud or
banks with valuable insights into customer dissatisfaction.
behaviour, needs, and preferences, as well as
market trends, risks, and opportunities. By using Sentiment analytics and big data analytics are
big data analytics techniques, such as machine transforming the banking sector by enabling
learning, clustering, and classification, banks can banks to become more customer-centric, agile,
perform customer segmentation more effectively and competitive. Banks that leverage these
and efficiently. They can identify patterns and technologies can gain a competitive edge in the
trends in customer data, segment customers into market and achieve higher profitability and
meaningful groups based on various criteria, such growth.
as demographics, lifestyle, spending habits,
channel preferences, and profitability, and create 5) Examine Customer Feedback: One of the
personalized offers and recommendations for each challenges faced by the banking sector is how to
segment. Customer segmentation using big data examine customer feedback and use it to improve
can help banks to improve customer retention and their products and services. Customer feedback
acquisition, increase cross-selling and up-selling can be collected from various sources, such as
opportunities, optimize pricing and promotions, surveys, reviews, social media, and online
enhance customer experience and satisfaction, and transactions. However, analysing this large and
reduce costs and risks. diverse data set requires advanced techniques and
4) Sentiment Analytics: Sentiment analytics and big tools. This is where big data comes in.
data analytics can be combined to create a Big data can help banks to examine customer
powerful tool for the banking sector. By applying feedback in a more efficient and effective way.
sentiment analytics to big data, banks can gain a For example, big data can help banks to:
deeper and more comprehensive understanding of
their customers' behaviours, attitudes, and - Identify customer needs, preferences, and
feedbacks across different channels and platforms. expectations
This can help banks to: - Segment customers based on their behaviour,
demographics, and preferences
- Enhance customer loyalty and retention by - Monitor customer satisfaction and loyalty
providing personalized and tailored products and - Detect customer complaints and issues
services that meet or exceed customer - Respond to customer feedback in a timely and
expectations. personalized manner
- Increase customer acquisition and cross-selling - Enhance customer experience and retention
by identifying potential customers and offering - Develop new products and services that meet
them relevant and attractive products and services customer demands
based on their needs and preferences.
- Reduce customer churn and complaints by By using big data to examine customer feedback,
detecting and resolving customer issues and banks can gain valuable insights into their
problems in a timely and effective manner.
 customers and improve their performance and
competitiveness in the market.
IX. FUTURE OF BIG DATA IN BANK SECTOR

VIII. DATA ANALYSIS As you can see, there is a wide range of applications
Results and Findings for big data in banking. All their endeavours, however,
During the internship, the data was analysed in one of merely scraped the surface.
the multinational banks. Questions were posed to bank The full potential of big data must also be realised in
employees working in IT divisions. banking.
1. Do you use Big Dat analytics for your banking According to Global Transaction Banking's
operations? whitepaper, 62% of banks feel big data is critical to
 Yes – 63%, No- 37% their success.
2. What are the main purposes for which you use big Despite this, just 29% of them believe their outcomes
data analytics? are of adequate business value.
 Risk management – 29%, Fraud detection Banks must rethink their operations and adopt data-
- 17%, CRM – 22%, New customer driven techniques to remain relevant and competitive.
acquisition – 15%, Social media Furthermore, big data will assist you in developing
Analytics-10% and expanding your banking firm.
3. How well trained are your personnel to leverage
the power of big data analytics?
CONCLUSION
X.
 Going through learning Curve - 52%,
High Level of expertise- 37%, Not very This study can be expanded to try and quantify the
sure – 11% financial and non-financial benefits that the banking
4. What are the main challenges that you face in sector received after implementing big data analytics, as
utilization of big data analytics? well as predict improvements in the bank's financial
 Data Analysis – 43%. Data warehousing – statements. This work can also be expanded to cover the
19%, Data Acquisition - 11%, Others – various data mining techniques that banks can use to
12%, Data visualization – 15% improve the quality of their analyses.
5. What is the importance of big data analytics for
formulating strategies in your banking operations? XI. REFERENCES
 Very Important – 59%, Important – 30%,
Not so important – 11%  https://www.mckinsey.com/industries/financial-
6. What benefits are you deriving by using big data services/our-insights/smarter-analytics-for-banks
analytics?  https://www.sciencedirect.com/science/article/pii/
 Better customer management – 27%, S1877050915005992
 https://www.simplilearn.com/data-science-ai-role-in-
Increase in business volume – 25%, banking-finance-article
Increase in profits – 21%, Gaining  https://bing.com/search?q=what+is+big+data
competitive edge – 17%, Others – 10%  https://www.investopedia.com/terms/b/big-data.asp
7. Has big data analytics been effective to provide  https://www.oracle.com/big-data/what-is-big-data/
meaningful insights into customer behaviour?  https://www.geeksforgeeks.org/what-is-structured-data/
 Yes – 77%, No – 23%  https://www.ibm.com/cloud/blog/structured-vs-
unstructured-data
8. What do you think will be the future drivers of
 https://www.tibco.com/reference-center/what-is-
banking operations? structured-data
 Mobile Banking – 29%, Near Field  https://www.geeksforgeeks.org/what-is-semi-
Communication (NFC) – 23%, remote structured-data/
banking - 21%, Internet Banking – 16%,  https://www.snowflake.com/guides/semi-structured-
Others – 11% data-101
 https://en.wikipedia.org/wiki/Semi-structured_data
 https://www.indeed.com/career-advice/career-
development/semi-structured-data
 https://monkeylearn.com/blog/semi-structured-data/
 https://www.techtarget.com/whatis/feature/A-history-
and-timeline-of-big-data
 https://www.dataversity.net/brief-history-big-data/
 https://www.bigdataframework.org/short-history-of-
big-data/
 https://link.springer.com/chapter/10.1007/978-3-319-
06245-7_3
 https://www.techtarget.com/searchdatamanagement/
feature/Top-trends-in-big-data-for-2021-and-beyond
 https://www.techtarget.com/whatis/feature/A-history-
and-timeline-of-big-data
 https://bing.com/search?
q=application+of+big+data+in+banking+sector
 https://www.analyticsinsight.net/how-big-data-is-
transforming-the-banking-sector/
 https://www.researchgate.net/publication/
357152903_Big_Data_Applications_the_Banking_Sect
or_A_Bibliometric_Analysis_Approach
 https://journals.sagepub.com/doi/full/
10.1177/21582440211067234
 https://www.analyticssteps.com/blogs/big-data-
banking-industry-benefits-uses-and-challenges
 https://www.irjet.net/archives/V7/i6/IRJET-
V7I61197.pdf
