Data Mining & Aggregation
Data mining, or Knowledge Discovery, as they say is a niche solution at Zivanta. Data mining and warehousing forms the crux of Zivanta’s backend operations. At Zivanta, we aggregate information from innumerable disparate data sources, using which we cull raw data useful for research and analysis in our projects and products. These raw data is first cleansed and structured in a manner that would be relevant to our research context, application development and database systems.
Our data mining experts cull data from sources depending on the project requirements. For example, in a recently completed strategy consulting project in the real estate sector, we dealt with enormous amounts of data from the U.S Census, Corelogic Mortgage Systems, Real Estate broking portals Zillow and Realtor, etc. With such humongous heterogeneous datasets it was imperative to understand what data we need and how. Our analysts used our mining algorithms to mine the required and relevant data to our server systems, structured the unstructured information so as to enable further analytics depending on the business requirements.
Likewise, in another mortgage hedge fund project where we had to deal with big data at loan level in the housing mortgage space, we employed mining strategies to develop process models to download, structure and analyze the data.
We also use data mining and text mining in our financial services products, like the credit card advisory portal and the deposit rate analytics portal.
Further to data mining, we also have in place our proprietary aggregation algorithms that we use in our crawling and scraping tools. We have further automated the aggregation process to streamline and reengineer business process and improve productivity.