Semantics of Ranking Queries for Probabilistic Data and Expected Ranks

Cormode, Graham; Li, Feifei; Yi, Ke

doi:10.1109/icde.2009.75

Cited by 161 publications

(217 citation statements)

References 29 publications

Supporting

Mentioning

214

Contrasting

Unclassified

Order By: Relevance

“…A lot of work was performed in the direction of ranking among uncertain data [3][4][5][6][7], but there is limited research on the inverse variant of ranking uncertain data [1]. In a nutshell, there are two models for capturing uncertainty of objects in a multi-dimensional space.…”

Section: Related Workmentioning

confidence: 99%

Continuous Inverse Ranking Queries in Uncertain Streams

Bernecker

Kriegel

Mamoulis

et al. 2011

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. This paper introduces a scalable approach for continuous inverse ranking on uncertain streams. An uncertain stream is a stream of object instances with confidences, e.g. observed positions of moving objects derived from a sensor. The confidence value assigned to each instance reflects the likelihood that the instance conforms with the current true object state. The inverse ranking query retrieves the rank of a given query object according to a given score function. In this paper we present a framework that is able to update the query result very efficiently, as the stream provides new observations of the objects. We will theoretically and experimentally show that the query update can be performed in linear time complexity. We conduct an experimental evaluation on synthetic and real-world data, which demonstrates the efficiency of our approach.

show abstract

Section: Related Workmentioning

confidence: 99%

Continuous Inverse Ranking Queries in Uncertain Streams

Bernecker

Kriegel

Mamoulis

et al. 2011

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…A number of probabilistic models have been proposed [3], [5], [7], [20]. Recently, a new model based on expected ranks has been developed in [10] to rank top-K uncertain objects. As shown in Section I, they may not always be sensitive to the relative distributions among object instances.…”

Section: Related Workmentioning

confidence: 99%

“…Choosing l = n 2 , the score distribution in Figure 1(a) is eventually modified to the score distribution in 1(c) after be insensitive to relative distributions of object instances. Very recently, in [10] a novel model based on the expected rank for ranking top-k uncertain objects has been proposed. Regarding the distributions and the permuted intermediate distributions as depicted above in Figures 1 (a)-(c), player A and B always have the same expected rank.…”

Section: Introductionmentioning

confidence: 99%

Quantile-based KNN over multi-valued objects

Zhang

Lin

Cheema

et al. 2010

2010 IEEE 26th International Conference on Data Engineering (ICDE 2010)

View full text Add to dashboard Cite

Abstract-K Nearest Neighbor search has many applications including data mining, multi-media, image processing, and monitoring moving objects. In this paper, we study the problem of KNN over multi-valued objects. We aim to provide effective and efficient techniques to identify KNN sensitive to relative distributions of objects. We propose to use quantiles to summarize relative-distribution-sensitive K nearest neighbors. Given a query Q and a quantile φ ∈ (0, 1], we firstly study the problem of efficiently computing K nearest objects based on a φ-quantile distance (e.g. median distance) from each object to Q. The second problem is to retrieve the K nearest objects to Q based on overall distances in the "best population" (with a given size specified by φ-quantile) for each object. While the first problem can be solved in polynomial time, we show that the 2nd problem is NP-hard. A set of efficient, novel algorithms have been proposed to give an exact solution for the first problem and an approximate solution for the second problem with the approximation ratio 2. Extensive experiment demonstrates that our techniques are very efficient and effective.

show abstract

“…{t 11 } is a possible world with probability p 11 (1 − p 21 In this letter, we focus on top k x-tuples detection using possible world semantics. Several queries in uncertain space have been proposed, such as U-topk, U-kRank, PT -k, Global-topk, Expected rank and so on [6]. In these queries, every tuple has a score and tuples with largest highrank probabilities would be returned.…”

Section: Introductionmentioning

confidence: 99%

Top (k1,k2) Query in Uncertain Datasets

Liu

Lin

Jia

2015

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYIn this letter, we propose a novel kind of uncertain query, top (k 1 , k 2 ) query. The x-tuple model and the possible world semantics are used to describe data objects in uncertain datasets. The top (k 1 , k 2 ) query is going to find k 2 x-tuples with largest probabilities to be the result of top k 1 query in a possible world. Firstly, we design a basic algorithm for top (k 1 , k 2 ) query based on dynamic programming. And then some pruning strategies are designed to improve its efficiency. An improved initialization method is proposed for further acceleration. Experiments in real and synthetic datasets prove the performance of our methods. key words: uncertain query, top k, x-tuple, possible world

show abstract

Semantics of Ranking Queries for Probabilistic Data and Expected Ranks

Cited by 161 publications

References 29 publications

Continuous Inverse Ranking Queries in Uncertain Streams

Continuous Inverse Ranking Queries in Uncertain Streams

Quantile-based KNN over multi-valued objects

Top (<i>k</i><sub>1</sub>,<i>k</i><sub>2</sub>) Query in Uncertain Datasets

Contact Info

Product

Resources

About