An interactive interface for multi-dimensional data stream analysis

Nuno C. Marques, Bruno Silva, Hugo Santos

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Data mining models are frequently used to represent and summarize meaningful properties in data. However such models are usually not suitable for interactive data exploration and visualization. This paper proposes the use of multidimensional projection together with the Ubiquitous Self-Organizing Map algorithm (UbiSOM), a novel variant of the well-known self-organizing map algorithm that was specifically tuned for stream data-mining. The resulting high-dimensional projection system is then studied for interactive data analysis and visualization. A prototype was developed where, at each moment, the user can visualize the information from different perspectives. Direct interaction with the system during stream processing is possible both by changing the projection, by optimizing the projection view for maximizing variance or by filtering the incoming data series. Experiments in two distinct datasets show the importance and relevance of conjoining multidimensional data projection with UbiSOM.

Original languageEnglish
Title of host publicationProceedings - 2016 20th International Conference Information Visualisation, IV 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages223-229
Number of pages7
Volume2016-August
ISBN (Electronic)9781467389426
DOIs
Publication statusPublished - 31 Aug 2016
Event20th International Conference Information Visualisation, IV 2016 - Lisbon, Portugal
Duration: 19 Jul 201622 Jul 2016

Publication series

NameIEEE International Conference on Information Visualization
PublisherIEEE
ISSN (Electronic)1550-6037

Conference

Conference20th International Conference Information Visualisation, IV 2016
CountryPortugal
CityLisbon
Period19/07/1622/07/16

Fingerprint

Self organizing maps
Data mining
Projection systems
Data visualization
Visualization
Processing
Experiments

Keywords

  • Big Data
  • Data Analysis
  • Data Streams
  • Interactive Data Mining
  • Multidimensional Data Projection
  • Self-Organizing Map

Cite this

Marques, N. C., Silva, B., & Santos, H. (2016). An interactive interface for multi-dimensional data stream analysis. In Proceedings - 2016 20th International Conference Information Visualisation, IV 2016 (Vol. 2016-August, pp. 223-229). [7557931] (IEEE International Conference on Information Visualization). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IV.2016.72
Marques, Nuno C. ; Silva, Bruno ; Santos, Hugo. / An interactive interface for multi-dimensional data stream analysis. Proceedings - 2016 20th International Conference Information Visualisation, IV 2016. Vol. 2016-August Institute of Electrical and Electronics Engineers Inc., 2016. pp. 223-229 (IEEE International Conference on Information Visualization).
@inproceedings{2ef24f57e6fc4a56bf54f4622524c6fc,
title = "An interactive interface for multi-dimensional data stream analysis",
abstract = "Data mining models are frequently used to represent and summarize meaningful properties in data. However such models are usually not suitable for interactive data exploration and visualization. This paper proposes the use of multidimensional projection together with the Ubiquitous Self-Organizing Map algorithm (UbiSOM), a novel variant of the well-known self-organizing map algorithm that was specifically tuned for stream data-mining. The resulting high-dimensional projection system is then studied for interactive data analysis and visualization. A prototype was developed where, at each moment, the user can visualize the information from different perspectives. Direct interaction with the system during stream processing is possible both by changing the projection, by optimizing the projection view for maximizing variance or by filtering the incoming data series. Experiments in two distinct datasets show the importance and relevance of conjoining multidimensional data projection with UbiSOM.",
keywords = "Big Data, Data Analysis, Data Streams, Interactive Data Mining, Multidimensional Data Projection, Self-Organizing Map",
author = "Marques, {Nuno C.} and Bruno Silva and Hugo Santos",
note = "Sem PDF.",
year = "2016",
month = "8",
day = "31",
doi = "10.1109/IV.2016.72",
language = "English",
volume = "2016-August",
series = "IEEE International Conference on Information Visualization",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "223--229",
booktitle = "Proceedings - 2016 20th International Conference Information Visualisation, IV 2016",
address = "United States",

}

Marques, NC, Silva, B & Santos, H 2016, An interactive interface for multi-dimensional data stream analysis. in Proceedings - 2016 20th International Conference Information Visualisation, IV 2016. vol. 2016-August, 7557931, IEEE International Conference on Information Visualization, Institute of Electrical and Electronics Engineers Inc., pp. 223-229, 20th International Conference Information Visualisation, IV 2016, Lisbon, Portugal, 19/07/16. https://doi.org/10.1109/IV.2016.72

An interactive interface for multi-dimensional data stream analysis. / Marques, Nuno C.; Silva, Bruno; Santos, Hugo.

Proceedings - 2016 20th International Conference Information Visualisation, IV 2016. Vol. 2016-August Institute of Electrical and Electronics Engineers Inc., 2016. p. 223-229 7557931 (IEEE International Conference on Information Visualization).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - An interactive interface for multi-dimensional data stream analysis

AU - Marques, Nuno C.

AU - Silva, Bruno

AU - Santos, Hugo

N1 - Sem PDF.

PY - 2016/8/31

Y1 - 2016/8/31

N2 - Data mining models are frequently used to represent and summarize meaningful properties in data. However such models are usually not suitable for interactive data exploration and visualization. This paper proposes the use of multidimensional projection together with the Ubiquitous Self-Organizing Map algorithm (UbiSOM), a novel variant of the well-known self-organizing map algorithm that was specifically tuned for stream data-mining. The resulting high-dimensional projection system is then studied for interactive data analysis and visualization. A prototype was developed where, at each moment, the user can visualize the information from different perspectives. Direct interaction with the system during stream processing is possible both by changing the projection, by optimizing the projection view for maximizing variance or by filtering the incoming data series. Experiments in two distinct datasets show the importance and relevance of conjoining multidimensional data projection with UbiSOM.

AB - Data mining models are frequently used to represent and summarize meaningful properties in data. However such models are usually not suitable for interactive data exploration and visualization. This paper proposes the use of multidimensional projection together with the Ubiquitous Self-Organizing Map algorithm (UbiSOM), a novel variant of the well-known self-organizing map algorithm that was specifically tuned for stream data-mining. The resulting high-dimensional projection system is then studied for interactive data analysis and visualization. A prototype was developed where, at each moment, the user can visualize the information from different perspectives. Direct interaction with the system during stream processing is possible both by changing the projection, by optimizing the projection view for maximizing variance or by filtering the incoming data series. Experiments in two distinct datasets show the importance and relevance of conjoining multidimensional data projection with UbiSOM.

KW - Big Data

KW - Data Analysis

KW - Data Streams

KW - Interactive Data Mining

KW - Multidimensional Data Projection

KW - Self-Organizing Map

UR - http://www.scopus.com/inward/record.url?scp=84989900379&partnerID=8YFLogxK

U2 - 10.1109/IV.2016.72

DO - 10.1109/IV.2016.72

M3 - Conference contribution

VL - 2016-August

T3 - IEEE International Conference on Information Visualization

SP - 223

EP - 229

BT - Proceedings - 2016 20th International Conference Information Visualisation, IV 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Marques NC, Silva B, Santos H. An interactive interface for multi-dimensional data stream analysis. In Proceedings - 2016 20th International Conference Information Visualisation, IV 2016. Vol. 2016-August. Institute of Electrical and Electronics Engineers Inc. 2016. p. 223-229. 7557931. (IEEE International Conference on Information Visualization). https://doi.org/10.1109/IV.2016.72