HEADLINES
JOBS
COM:PORT
POLICY POLICE
ARCHIVES

Focus rests on data warehousing and mining

The first track on data and security was delivered by Gaurav Verma, product manager, Oracle India, and Gourish Hosangady, national technical manager, SAS Institute (India).

With the focus on data warehousing processes, Verma's talk took the audience through what really is involved here.

Data warehouse comprises technology components that allow access to data from multiple services and to manage and analyse that data.

The complete solution involves design, extraction, building, managing and analysing.

The implementation approach is more a subject of business and IT strategy. The important aspect is to do a proper study of the scope for services and important deliverables.

Technical architecture components are required to have more and more modelling and less of coding. The important features involve leveraging RDBMS technology for scalability and high performance and support for ERP applications data.

The integration of technologies requires a decision on increments: OLAP (online analytical processing) or data mining.

The business phase requires one to look at technology architecture, increments and infrastructure.

Hosangady elaborated on data mining. The objective of data warehousing and data mining is to enable every worker to be a 'knowledge worker' by reaching the required data.

It allows for an enterprise-wide 'customer view'. Data Mart, a small-scale version of data warehousing, OLAP, data mining and data warehousing constitutes possible enterprise solutions.

While OLAP is more useful for knowing what is happening as a clue to charge strategies, data mining tells why things are happening and enable deployment of preventive measures.

The primary purpose of data mining is to better understand the customer through selection of data from large amounts of date, for business advantage.

Customer-relations management is becoming very important today and all sections of the industry and even governments now need data mining for achieving this.

Banks, with the large 'non-paying assets would benefit most by data mining as they would learn to admit customer credibility and risks.

Business requirements demand that there be better marketing, segmentation and profiling, targeted marketing and management of attrition or return of customers.

Since it costs about eight times as much to attract a new customer as it does to retain an existing one, CRM is a huge initiative currently.

Traditional statistics, data visualisation, clustering, classification and decision trees, artificial neural networks, are tools for data mining.

The business-customer relations are enabled by data mining so as to enhance acquisition, cross-selling, up-selling, retention and winning back of customers.

Customer loyalty is defined as the reason why people stay with an organisation apart from the price, quality and other measurable attributes. Data mining gives valuable clues for retaining customers loyalty.

The major barriers to data mining are the lack of data, expenses, multiple disconnected sources, organisational, legal factors etc.

"Undigested data in like idle money. Mining terms this into profit. Reach the customers that count, don't cannot the customers you reach!" he concluded.

Tell us what you think