**2. Literature review**

Many of today's query processing platforms carry a much profound repertoire and resilient querying techniques to regulate huge observational data in a limited resource environment. [12, 11]. Below some aspects are reviewed that helps user in searching relevant information from a large data set. We consider some prominent researches delivered for automatic exploration in data space, formulation of approximate quires and techniques that assist the user in the query formulation process to cover our aspects.

## **2.1. Automatic exploration**

Analyzing vast amount of real time data can be an extremely complex task and required automation. In such scenario without any assistance, user ended up with ill-formulated query that retrieves no result or huge result set. Traditional Database Management tools and systems are constructed by considering that database semantics is well understood by users [39]. Therefore, current applications with huge and complex database do not work well with these traditional Data Base Management techniques. Many interactive data exploration strategies are proposed and developed by researchers that extract and uncover great knowledge from complex data via highly ad-hoc interaction.

Automatic Interactive Data Exploration (AIDE) framework is well explained in [16] by authors. In that, the user is directed towards the data area of interest by deliberately incorporating relevance feedback. Various machine learning and data mining techniques can be integrated in that to achieve the best performance. Similarly, in [17] YAML framework is suggested and it uses attribute-value pair frequency to make exploration effective. Automatic exploration strategy performs formulation of user's queries and leads towards relevant information.
