Efficient sampling strategies for relational database operations
ICDT Selected papers of the 4th international conference on Database theory
Query size estimation by adaptive sampling
Selected papers of the 9th annual ACM SIGACT-SIGMOD-SIGART symposium on Principles of database systems
Scaling Up a Boosting-Based Learner via Adaptive Sampling
PADKK '00 Proceedings of the 4th Pacific-Asia Conference on Knowledge Discovery and Data Mining, Current Issues and New Applications
Adaptive Sampling Methods for Scaling Up Knowledge Discovery Algorithms
DS '99 Proceedings of the Second International Conference on Discovery Science
Analysis and application of adaptive sampling
Journal of Computer and System Sciences - Special issue on PODS 2000
Sequential sampling techniques for algorithmic learning theory
Theoretical Computer Science - Algorithmic learning theory (ALT 2000)
Hi-index | 0.00 |
Sampling is an important technique for parameter estimation and hypothesis testing widely used in statistical analysis, machine learning and knowledge discovery. In contrast to batch sampling methods in which the number of samples is known in advance, adaptive sequential sampling gets samples one by one in an on-line fashion without a pre-defined sample size. The stopping condition in such adaptive sampling scheme is dynamically determined by the random samples seen so far. In this paper, we present a new adaptive sequential sampling method for estimating the mean of a Bernoulli random variable. We define the termination conditions for controlling the absolute and relative errors. We also briefly present a preliminary theoretical analysis of the proposed sampling method. Empirical simulation results show that our method often uses significantly lower sample size (i.e., the number of sampled instances) while maintaining competitive accuracy and confidence when compared with most existing methods such as that in [14]. Although the theoretical validity of the sampling method is only partially established. we strongly believe that our method should be sound in providing a rigorous guarantee that the estimation results under our scheme have desired accuracy and confidence.