Summary

The growth of organizations invariably leads to creation of multiple isolated data sources which are totally disconnected from each other. This leads to reduced efficiency and lack of complete knowledge of the enterprise and its customers while making critical business decisions. This is the classical Information integration problem which has become the biggest pain point for enterprises today. Information Integration has received considerable attention from researchers in academia as well as industry in the recent past. Information integration refers to the category of middleware which lets applications access data as though there were in a single database. The popularity of information integration has lead to it being used in diverse domains. Each domain has its peculiar requirements; like in some domains it is not feasible to know the schema of all the data sources whereas in some domains the information integration has to be aware of the users’ access rights. Such requirements lead to new approaches for doing information integration. In this talk, we will discuss new trends and techniques for integrating information that provide organizations the ability to generate insights that would not be normally obtained by analyzing either type of information source (structured or unstructured) independently. We further show that this new way of integrating information has several applications in the area of active data warehousing, semantic search, and business intelligence to provide real-time and personalized analytics in the CRM domain.