Page 123 - Special Topic Session (STS) - Volume 1
P. 123
STS419 Liza M. et al.
networks and the internet of things (Cornelia, L.H., Diane, C.K., and Gabriel, Q.
2017). Despite the lack of consensus on the definition of Big Data, its
characteristics are commonly associated with the 3V’s (Cornelia, L.H., et al.
2017; Devakunchari, R. 2014; Kshetri, N. 2014) which are Volume (Large data
size in terabytes and petabytes), Velocity (The rate at which data flows in from
sources), and Variety (Structured, semi-structured and unstructured data).
Despite the extensive existing research on Big Data, limited literature
observes Big Data’s impact on social finance for Islamic Financial Institutions.
This paper attempts to deepen the literature by developing a descriptive
framework for Islamic Financial Institutions to leverage off Big Data for social
finance. Our contribution is to foster a better understanding of the possibilities
behind Big Data’s for sustainable and social finance. Thus, providing a basis
for future studies on the areas where Islamic Financial intermediaries could
use Big Data to better coordinate and promote sustainable and social finance.
2. Methodology
This study proposes a method that could be used to utilise financial data
to develop a predictive model for cash Waqf beneficiaries. The data
recommended for this would be the Islamic financial intermediary’s customer
information and financial history. The time period observed is suggested to
span long periods however for the purpose of this paper the illustrative
example will cover a one month period. The findings would be used as a basis
for machine learning to develop a predictive model via a decision tree model.
The following section sets out a significant part of this study which is
qualitative, conceptual and aimed at building a theoretical framework.
3. Analysis
In order to gain insight into the potential characteristics that would be
applicable to Big Data analytics, we had to first identify the relevant
information. In accordance with the identified area of focus, the span of data
was determined to a duration of a month. However, the decision to limit the
span of data impacted the available sample size, which is considered in the
observation of the results.
The primary data was the biproduct of the core business, which is readily
available information captured to extend its services. Due to the continuous
update process stemming from its business activities, it allows for a live study
of customer behavior. On the spectrum of data types which consists of
Unstructured, Multi-Structured, and Structured data, the data available would
be mostly derived from structured data.
The data is first collected via forms followed by the information input of
data into groups that will be key in assessing demographics. The vast amounts
of data are then pooled into a data warehouse where each data set is assigned
112 | I S I W S C 2 0 1 9