Predictive Analytics in Financial Service

This article explores the use of predictive analytics in the financial services industry, including its applications in fraud detection, credit risk measurement, and customer analysis. It discusses the use of predictive modeling to identify unknown patterns and forecast future events based on historical data. The article also provides a case study of how PayPal used predictive analytics to analyze customer feedback and improve their products. Overall, predictive analytics is a powerful tool for financial services to solve problems and uncover new opportunities.

Running head: Predictive analytics in financial service
Name:
Institution affiliation:
Subject:
Date:

Predictive analytic in financial service 1.
According to (Bharadwaj, 2019) nowadays, the business has implemented automation in
its operations. This event has resulted in the collection of a lot of data. It is therefore evident that
analytic data techniques are growing exponentially in business sectors. Predictive analytics and
predictive modeling are one of the most-known analytic tools used.
(FIRMANI, 2019) Defines predictive analytics as to the use of machine learning and
statistical analyzing techniques to predict the occurrence of a particular event in future form set
of historical data. He continues to add that business nowadays leverage predictive modeling to
discover complex data correlations to identify unknown patterns and well as forecasting.
Predictive analytics and modeling have gained its routes in different industries and
business sectors. For this matter, we wish to present a use case of the technique in financial
services. Predictive analytics helps in solving problems and helps in unleashing new
opportunities. In commercial sectors, for instance, banks, predictive analytics have been
embraced to detect fraud as well as reducing it. It has its roots in the measurement of credit risk.
It is, however, worth noting that data form financial services like banks tend to form a certain
Patten. For instance, the data can show a lot of significant transaction during the night, and this
might suggest that there might be some of the rules, for example, is evaded during the night. It is
therefore reasonable to uses pattern tracking as the best data mining technique in this form of
industry. In banking, for instance, data mining helps in: (1) forecasting the most probable loan
defaulters,(2) detection online banking and fraudulent and (4) identifying the potentially
profitable customers and their preferred needs. (Sharda, 2011) . An excellent example of a
finance case application of the predictive analytics and predictive model is Rapid miner
(Ristoski, 2015). The company was founded in 2007 ever since then; it has been building
software to do data analysis. Some of the predictive analytic that Rapidminer offers includes;

Predictive analytic in financial service 2.
demand forecasting, fraud detection, and preventing churn among others. Recently PayPal is
working with the company. (Verma, 2014) Claims that predictive analytics software by
Rapidminer is capable of ;( 1) monitoring all transaction from the customers hence preventing
any fraudulent transaction. (2) use of historical promotion data to forecast the impact of customer
engagement in the direct market promotion.
There is a claim that Rapidminer can help the business achieve the above two goals by
leveraging their historical data. For instance, in case of prediction of customer impact on the
engagement on the retail shop. The company (Rapidminer) will first work in hand with the
retailer to get all the transaction data and all historical data on the transaction for a particular
product. Then data will be filtered into a model that is usable in machine learning. Mathematical
operations such as regression and analysis of variance are then performed to identify any pattern
from the data. From the model, we are able now to investigate the effect of changing one
variable to the other.
PayPal an international online money transaction company engaged with Rapidminer to
gauge the tension of their top customers. It also wished to monitor their complaints as well. The
reason behind this engagement is that PayPal wanted to analyze their customer feedback to get a
better understanding of drivers of product improvement.
PayPal had a challenge with the task to analyze its customers' comment. Thanks to
Rapidminer, which leveraged on artificial intelligence and the PayPal data engineers who joined
hands to develop a system. This system could perform a sentiment analysis of the customers'
comment, from all its social media (Kagan, 2013).

Predictive analytic in financial service 3.
According to articles by Nonprofit Business Advisor in 2016 Rapidminer was the first to
extract the frequently used world by the PayPal customers when complaining. It also identified
the top login related issues, which then helped PayPal strategized a solution to these issues.
Rapidminer, with the help of predictive analytics, was able to locate that most of the password
issues arose during November and December. Rapid claims that it helped PayPal to structure the
permanent remedy to this login issues, which resulted in a 50% success login after 2-3 week of
the investigation.
We can implement various predictive analytic strategies in this uses case. Nevertheless,
in this specific financial service use case, social network analysis works best. It is because it will
provide a link between the actors and ties in financial matters (Traymbak, 2015). Actors refer to
the human participant, while ties relate to the connections between the people. To leverage on
the historical data to predict future frauds, for instance, linear regression modeling will be used.
This event will result in an analysis that will show the effect of one or two independent variables
in the financial sector to the depended variable (Zhou, 2005). For the cases of PayPal, a decision,
for example, is much useful. It provided the company with information on which type of
comment to put emphasis more than others. As shown above, it is clear that predictive analytic
aims at identifying unknown future events based on historical data, statistical modeling, and
artificial intelligence.

Predictive analytic in financial service 4.
References
Bharadwaj, R. (2019, May 19). Predictive Analytics – 5 Examples of Industry Applications. Retrieved from
emerj.com: https://emerj.com/ai-sector-overviews/predictive-analytics-5-examples-of-industry-
applications/
FIRMANI, F. (2019, May 12). 6 Ways Predictive Analytics in Insurance will Shape the Industry in 2019.
Retrieved from www.duckcreek.com: https://www.duckcreek.com/blog/predictive-analyitics-
reshaping-insurance-industry/
Kagan, V. (2013). Sentiment Analysis for PTSD Signals. New York, NY: Springer.
PayPal reports record end-of-year charitable giving. ( 2016). Nonprofit Business Advisor, 7-8.
Ristoski, P. (2015). Mining the Web of Linked Data with Rapidminer. SSRN Electronic Journal.
Sharda, R. (2011, January 4). Decision Making and Analytics. Retrieved from seu1.org:
http://seu1.org/files/level8/IT445/IT445%20BOOK%20EDIT.pdf
Traymbak, S. (2015). A Collaborative Approach to Managerial Decision Making Through Integration of
Data Mining and Predictive Analytic. SSRN Electronic Journal.
Verma, T. (2014). Tokenization and Filtering Process in RapidMiner. International Journal of Applied
Information Systems, 16-18.
Zhou, Y. (2005). IEEE Intelligent Systems. US Domestic Extremist Groups on the Web: Link and Content
Analysis, 44-51.