In recent years, the Indian Economy has shown tremendous
growth. Due to advances in data storage and networking technologies,
large amounts of data are now available. Data mining techniques
can be applied to these large volumes of data to come up
with meaningful information that can help in making businesses
more competitive and sustain economic development in the
long run. The question arises as to what kind of infrastructural
setup would be appropriate to perform data mining operations
across various industry segments in a developing country
like India. This paper proposes two kinds of infrastructural
setups namely centralized and decentralized setups for performing
data mining operations. It focuses on three specific areas
of potential applications of data mining in the growth of
the Indian economy—analysis of risk in finance and
banking sector, retail business, health insurance and fraud
management—and suggests which kind of infrastructure
may be suitable for each specific application area.
With the recent advances in the fields of computing, communications,
and digital storage technologies and also the developments
in throughput data acquisition technologies, it is now possible
to gather and store huge volumes of data. Warehouses nowadays
are capable of storing massive quantities of information
about multiple aspects of business operations. Warehouses
of large retailers both international (like Walmart) and
national (like Big Bazaar, Pantaloons, etc.), are typically
multi-terabyte databases that can store information on retail
transactions done by the customers in distributed geographical
regions within the same country or all over the world. With
technological advances, it is now possible to have complex
distributed systems consisting of computer systems, communication
networks and power systems to be equipped with sensors and
measurement devices that can gather and store different
kinds of data to be used in monitoring, controlling and
improving operations (Pottie and Kaiser, 2000).
|
