Adaptive-Size Reservoir Sampling over Data Streams

Al-Kateb, Mohammed; Lee, Byung Suk; Wang, X.S.

doi:10.1109/ssdbm.2007.29

Cited by 45 publications

(36 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then a randomly chosen item is discarded from the reservoir to free space for the new item if added. The ith item is discarded with probability 1 − p. A number of sampling techniques have been investigated like in [Al-Kateb et al 2007;Efraimidis and Spirakis 2006;Aggarwal 2006;Rusu and Dobra 2009]. Reservoir sampling has been discussed in some studies related to drift, change detection and as an alternative to windowing in [Ng and Dash 2008;Yao et al 2012;Zhao et al 2011].…”

Section: Memorymentioning

confidence: 99%

A survey on concept drift adaptation

et al. 2014

View full text Add to dashboard Cite

Concept drift primarily refers to an online supervised learning scenario when the relation between the input data and the target variable changes over time. Assuming a general knowledge of supervised learning in this paper we characterize adaptive learning process, categorize existing strategies for handling concept drift, discuss the most representative, distinct and popular techniques and algorithms, discuss evaluation methodology of adaptive algorithms, and present a set of illustrative applications. This introduction to the concept drift adaptation presents the state of the art techniques and a collection of benchmarks for researchers, industry analysts and practitioners. The survey aims at covering the different facets of concept drift in an integrated way to reflect on the existing scattered state-of-the-art.

show abstract

Section: Memorymentioning

confidence: 99%

A survey on concept drift adaptation

et al. 2014

View full text Add to dashboard Cite

show abstract

“…Then, we conduct a theoretical study on the effects of increasing a reservoir size in the middle of sampling, and propose our progressive reservoir sampling algorithm. Results of this theoretical study and the origin of the proposed algorithm appear in [3].…”

Section: Progressive Reservoir Samplingmentioning

confidence: 99%

“…In [20], Johnson et al abstract the process of sampling from a stream and design a generic sampling operator which can be specialized to implement a wide variety of stream sampling algorithms. In [3], we theoretically study the effects of increasing and decreasing the reservoir size in the middle of sampling on the sample uniformity. We present an adaptive-size reservoir sampling algorithm and apply it to an application in which samples are collected from wireless sensor networks using a mobile sink.…”

Section: Related Workmentioning

confidence: 99%

“…For this, we modify the conventional reservoir sampling technique to what we call the progressive reservoir sampling [3]. The key idea of PRJS is to exploit the property of reservoir sampling that the sampling probability keeps decreasing for each subsequent tuple.…”

Section: Introductionmentioning

confidence: 99%

“…Once the reservoir size is increased, the sample's uniformity is damaged. Besides, even after the enlarged reservoir is filled again with new tuples, the sample's uniformity is still not guaranteed, that is, the sample's uniformity confidence stays below 100% [3] (details in Section 2). There is a tradeoff that a larger increase of reservoir size leads to lower uniformity confidence after the reservoir is filled again.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations