A Constrained Randomization Approach to Interactive Visual Data Exploration with Subjective Feedback

Bo Kang, Kai Puolamäki, Jefrey Lijffijt, Tijl de Bie

Tutkimustuotos: ArtikkelijulkaisuArtikkeliTieteellinenvertaisarvioitu

Kuvaus

Data visualization and iterative/interactive data mining are growing rapidly in attention, both in research as well as in industry. However, while there are plethora of advanced data mining methods and lots of works in the field of visualisation, integrated methods that combine advanced visualization and/or interaction with data mining techniques in a principled way are rare. We present a framework based on constrained randomization which lets users explore high-dimensional data via ‘subjectively informative’ two-dimensional data visualizations. The user is presented with ‘interesting’ projections, allowing users to express their observations using visual interactions that update a background model representing the user’s belief state. This background model is then considered by a projection-finding algorithm employing data randomization to compute a new ‘interesting’ projection. By providing users with information that contrasts with the background model, we maximize the chance that the user encounters striking new information present in the data. This process can be iterated until the user runs out of time or until the difference between the randomized and the real data is insignificant. We present two case studies, one controlled study on synthetic data and another on census data, using the proof-of-concept tool SIDE that demonstrates the presented framework.
Alkuperäiskielienglanti
LehtiIEEE Transactions on Knowledge and Data Engineering
Sivumäärä14
ISSN1041-4347
DOI - pysyväislinkit
TilaE-pub ahead of print - 18 huhtikuuta 2019
OKM-julkaisutyyppiA1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä, vertaisarvioitu

Tieteenalat

  • 113 Tietojenkäsittely- ja informaatiotieteet

Lainaa tätä

@article{145357c3e9814090bc88e26eb5e6e6e0,
title = "A Constrained Randomization Approach to Interactive Visual Data Exploration with Subjective Feedback",
abstract = "Data visualization and iterative/interactive data mining are growing rapidly in attention, both in research as well as in industry. However, while there are plethora of advanced data mining methods and lots of works in the field of visualisation, integrated methods that combine advanced visualization and/or interaction with data mining techniques in a principled way are rare. We present a framework based on constrained randomization which lets users explore high-dimensional data via ‘subjectively informative’ two-dimensional data visualizations. The user is presented with ‘interesting’ projections, allowing users to express their observations using visual interactions that update a background model representing the user’s belief state. This background model is then considered by a projection-finding algorithm employing data randomization to compute a new ‘interesting’ projection. By providing users with information that contrasts with the background model, we maximize the chance that the user encounters striking new information present in the data. This process can be iterated until the user runs out of time or until the difference between the randomized and the real data is insignificant. We present two case studies, one controlled study on synthetic data and another on census data, using the proof-of-concept tool SIDE that demonstrates the presented framework.",
keywords = "113 Computer and information sciences",
author = "Bo Kang and Kai Puolam{\"a}ki and Jefrey Lijffijt and Bie, {Tijl de}",
year = "2019",
month = "4",
day = "18",
doi = "10.1109/TKDE.2019.2907082",
language = "English",
journal = "IEEE Transactions on Knowledge and Data Engineering",
issn = "1041-4347",
publisher = "IEEE Computer Society",

}

A Constrained Randomization Approach to Interactive Visual Data Exploration with Subjective Feedback. / Kang, Bo; Puolamäki, Kai; Lijffijt, Jefrey; Bie, Tijl de.

julkaisussa: IEEE Transactions on Knowledge and Data Engineering, 18.04.2019.

Tutkimustuotos: ArtikkelijulkaisuArtikkeliTieteellinenvertaisarvioitu

TY - JOUR

T1 - A Constrained Randomization Approach to Interactive Visual Data Exploration with Subjective Feedback

AU - Kang, Bo

AU - Puolamäki, Kai

AU - Lijffijt, Jefrey

AU - Bie, Tijl de

PY - 2019/4/18

Y1 - 2019/4/18

N2 - Data visualization and iterative/interactive data mining are growing rapidly in attention, both in research as well as in industry. However, while there are plethora of advanced data mining methods and lots of works in the field of visualisation, integrated methods that combine advanced visualization and/or interaction with data mining techniques in a principled way are rare. We present a framework based on constrained randomization which lets users explore high-dimensional data via ‘subjectively informative’ two-dimensional data visualizations. The user is presented with ‘interesting’ projections, allowing users to express their observations using visual interactions that update a background model representing the user’s belief state. This background model is then considered by a projection-finding algorithm employing data randomization to compute a new ‘interesting’ projection. By providing users with information that contrasts with the background model, we maximize the chance that the user encounters striking new information present in the data. This process can be iterated until the user runs out of time or until the difference between the randomized and the real data is insignificant. We present two case studies, one controlled study on synthetic data and another on census data, using the proof-of-concept tool SIDE that demonstrates the presented framework.

AB - Data visualization and iterative/interactive data mining are growing rapidly in attention, both in research as well as in industry. However, while there are plethora of advanced data mining methods and lots of works in the field of visualisation, integrated methods that combine advanced visualization and/or interaction with data mining techniques in a principled way are rare. We present a framework based on constrained randomization which lets users explore high-dimensional data via ‘subjectively informative’ two-dimensional data visualizations. The user is presented with ‘interesting’ projections, allowing users to express their observations using visual interactions that update a background model representing the user’s belief state. This background model is then considered by a projection-finding algorithm employing data randomization to compute a new ‘interesting’ projection. By providing users with information that contrasts with the background model, we maximize the chance that the user encounters striking new information present in the data. This process can be iterated until the user runs out of time or until the difference between the randomized and the real data is insignificant. We present two case studies, one controlled study on synthetic data and another on census data, using the proof-of-concept tool SIDE that demonstrates the presented framework.

KW - 113 Computer and information sciences

U2 - 10.1109/TKDE.2019.2907082

DO - 10.1109/TKDE.2019.2907082

M3 - Article

JO - IEEE Transactions on Knowledge and Data Engineering

JF - IEEE Transactions on Knowledge and Data Engineering

SN - 1041-4347

ER -