Conference proceeding
Supporting Dynamic Quantization for High-Dimensional Data Analytics
Proceedings of the ExploreDB'17. International Workshop on Exploratory Search in Databases and the Web (4th : 2017 : Chicago, Ill.), Vol.2017, pp.1-6
05/2017
DOI: 10.1145/3077331.3077336
PMCID: PMC5947868
PMID: 29757329
Abstract
Similarity searches are at the heart of exploratory data analysis tasks. Distance metrics are typically used to characterize the similarity between data objects represented as feature vectors. However, when the dimensionality of the data increases and the number of features is large, traditional distance metrics fail to distinguish between the closest and furthest data points. Localized distance functions have been proposed as an alternative to traditional distance metrics. These functions only consider dimensions close to query to compute the distance/similarity. Furthermore, in order to enable interactive explorations of high-dimensional data, indexing support for ad-hoc queries is needed. In this work we set up to investigate whether bit-sliced indices can be used for exploratory analytics such as similarity searches and data clustering for high-dimensional big-data. We also propose a novel dynamic quantization called Query dependent Equi-Depth (QED) quantization and show its effectiveness on characterizing high-dimensional similarity. When applying QED we observe improvements in kNN classification accuracy over traditional distance functions.
Gheorghi Guzun and Guadalupe Canahuate. 2017. Supporting Dynamic Quantization for High-Dimensional Data Analytics. In Proceedings of Ex-ploreDB'17, Chicago, IL, USA, May 14-19, 2017, 6 pages. https://doi.org/http://dx.doi.org/10.1145/3077331.3077336.
Details
- Title: Subtitle
- Supporting Dynamic Quantization for High-Dimensional Data Analytics
- Creators
- Gheorghi Guzun - Electrical and Computer Engineering, University of Iowa, Iowa City, Iowa 52242Guadalupe Canahuate - Electrical and Computer Engineering, University of Iowa, Iowa City, Iowa 52242
- Resource Type
- Conference proceeding
- Publication Details
- Proceedings of the ExploreDB'17. International Workshop on Exploratory Search in Databases and the Web (4th : 2017 : Chicago, Ill.), Vol.2017, pp.1-6
- DOI
- 10.1145/3077331.3077336
- PMID
- 29757329
- PMCID
- PMC5947868
- Grant note
- name: NIH, award: R01CA214825; DOI: 10.13039/100000001, name: NSF, award: DMS-1557578
- Language
- English
- Date published
- 05/2017
- Academic Unit
- Electrical and Computer Engineering
- Record Identifier
- 9984083859202771
Metrics
23 Record Views