A large number of web data repositories are hidden behind restrictive web interfaces, making it an important challenge to enable data analytics over these hidden web databases. In this propose system novel techniques which use a small number of queries to produce unbiased estimates with small variance.This paper presents real datasets demonstrate the accuracy and efficiency algorithms. Left-deep-tree data structure which imposes an order of all queries. Based on the order, it is capable of mapping each tuple in the hidden database to exactly one query in the tree, which is referred as the designated query. In this propose system novel techniques which use a small number of queries to produce unbiased estimates with small variance. These techniques can also be used for approximate query processing over hidden databases. Present theoretical analysis and extensive experiments to illustrate the effectiveness of proposed approach. In this paper initiate a study of estimating, without bias, the size and other aggregates over a hidden database.