### Abstract

Nearest-neighbor queries, which ask for returning the nearest neighbor of a query point in a set of points, are important and widely studied in many fields because of a wide range of applications. In many of these applications, such as sensor databases, location based services, face recognition, and mobile data, the location of data is imprecise. We therefore study nearest neighbor queries in a probabilistic framework in which the location of each input point and/or query point is specified as a probability density function and the goal is to return the point that minimizes the expected distance, which we refer to as the expected nearest neighbor (ENN). We present methods for computing an exact ENN or an ε-approximate ENN, for a given error parameter 0 < ε < 1, under dierent distance functions. These methods build an index of near-linear size and answer ENN queries in polylogarithmic or sublinear time, depending on the underlying function. As far as we know, these are the first nontrivial methods for answering exact or ε-approximate ENN queries with provable performance guarantees.

Original language | English (US) |
---|---|

Title of host publication | Proceedings of the ACM SIGACT-SIGMOD-SIGART Symposium on Principles of Database Systems |

Pages | 225-236 |

Number of pages | 12 |

DOIs | |

State | Published - 2012 |

Event | 31st ACM SIGMOD-SIGACT-SIGART Symposium on Principles of Database Systems, PODS '12 - Scottsdale, AZ, United States Duration: May 21 2012 → May 23 2012 |

### Other

Other | 31st ACM SIGMOD-SIGACT-SIGART Symposium on Principles of Database Systems, PODS '12 |
---|---|

Country | United States |

City | Scottsdale, AZ |

Period | 5/21/12 → 5/23/12 |

### Fingerprint

### Keywords

- approximate nearest neighbor
- expected nearest neighbor (enn)
- indexing uncertain data
- nearest-neighbor queries

### ASJC Scopus subject areas

- Software
- Information Systems
- Hardware and Architecture

### Cite this

*Proceedings of the ACM SIGACT-SIGMOD-SIGART Symposium on Principles of Database Systems*(pp. 225-236) https://doi.org/10.1145/2213556.2213588

**Nearest-neighbor searching under uncertainty.** / Agarwal, Pankaj K.; Efrat, Alon; Sankararaman, Swaminathan; Zhang, Wuzhou.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

*Proceedings of the ACM SIGACT-SIGMOD-SIGART Symposium on Principles of Database Systems.*pp. 225-236, 31st ACM SIGMOD-SIGACT-SIGART Symposium on Principles of Database Systems, PODS '12, Scottsdale, AZ, United States, 5/21/12. https://doi.org/10.1145/2213556.2213588

}

TY - GEN

T1 - Nearest-neighbor searching under uncertainty

AU - Agarwal, Pankaj K.

AU - Efrat, Alon

AU - Sankararaman, Swaminathan

AU - Zhang, Wuzhou

PY - 2012

Y1 - 2012

N2 - Nearest-neighbor queries, which ask for returning the nearest neighbor of a query point in a set of points, are important and widely studied in many fields because of a wide range of applications. In many of these applications, such as sensor databases, location based services, face recognition, and mobile data, the location of data is imprecise. We therefore study nearest neighbor queries in a probabilistic framework in which the location of each input point and/or query point is specified as a probability density function and the goal is to return the point that minimizes the expected distance, which we refer to as the expected nearest neighbor (ENN). We present methods for computing an exact ENN or an ε-approximate ENN, for a given error parameter 0 < ε < 1, under dierent distance functions. These methods build an index of near-linear size and answer ENN queries in polylogarithmic or sublinear time, depending on the underlying function. As far as we know, these are the first nontrivial methods for answering exact or ε-approximate ENN queries with provable performance guarantees.

AB - Nearest-neighbor queries, which ask for returning the nearest neighbor of a query point in a set of points, are important and widely studied in many fields because of a wide range of applications. In many of these applications, such as sensor databases, location based services, face recognition, and mobile data, the location of data is imprecise. We therefore study nearest neighbor queries in a probabilistic framework in which the location of each input point and/or query point is specified as a probability density function and the goal is to return the point that minimizes the expected distance, which we refer to as the expected nearest neighbor (ENN). We present methods for computing an exact ENN or an ε-approximate ENN, for a given error parameter 0 < ε < 1, under dierent distance functions. These methods build an index of near-linear size and answer ENN queries in polylogarithmic or sublinear time, depending on the underlying function. As far as we know, these are the first nontrivial methods for answering exact or ε-approximate ENN queries with provable performance guarantees.

KW - approximate nearest neighbor

KW - expected nearest neighbor (enn)

KW - indexing uncertain data

KW - nearest-neighbor queries

UR - http://www.scopus.com/inward/record.url?scp=84862649949&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862649949&partnerID=8YFLogxK

U2 - 10.1145/2213556.2213588

DO - 10.1145/2213556.2213588

M3 - Conference contribution

SN - 9781450312486

SP - 225

EP - 236

BT - Proceedings of the ACM SIGACT-SIGMOD-SIGART Symposium on Principles of Database Systems

ER -