Visual analytics of educational time-dependent data using interactive dynamic visualization

GÉRYK, Jan. Visual analytics of educational time-dependent data using interactive dynamic visualization. Expert Systems. 2017, roč. 34, č. 1, s. 1-14. ISSN 0266-4720. Dostupné z: https://dx.doi.org/10.1111/exsy.12175.

Základní údaje

• Originální název: Visual analytics of educational time-dependent data using interactive dynamic visualization
• Autoři: GÉRYK, Jan (203 Česká republika, garant, domácí).
• Vydání: Expert Systems, 2017, 0266-4720.

Další údaje

• Originální jazyk: angličtina
• Typ výsledku: Článek v odborném periodiku
• Obor: 10201 Computer sciences, information science, bioinformatics
• Stát vydavatele: Velká Británie a Severní Irsko
• Utajení: není předmětem státního či obchodního tajemství
• WWW: URL
• Impakt faktor: Impact factor: 1.430
• Kód RIV: RIV/00216224:14330/17:00095878
• Organizační jednotka: Fakulta informatiky
• Doi: http://dx.doi.org/10.1111/exsy.12175
• UT WoS: 000394618100006
• Klíčová slova anglicky: animation; cluster analysis; motion charts; trajectory analysis; visual analytics
• Příznaky: Mezinárodní význam, Recenzováno

Anotace

A major challenge in the modern educational research is the large amount of time-dependent data, which requires suitable methods to provide an efficient decision making. The fields of visual analytics and interactive visualization can help to cope with the related issues by involving the analyst in the exploration process. The time is an important variable that needs to be modeled in visual analytics. Methods like motion charts show the changes over time by presenting animations in the two-dimensional space and by changing the element appearances. In this paper, we present a novel web-based visual analytics framework, which is primarily designed for exploratory analysis of academic analytics and employs a combination of improved motion charts methods, trajectory-based visualizations, and automated analytic methods for the purpose of analyzing data elements based on their relations with the time dimension and examining data dynamics. We evaluate the usefulness and the general applicability of the framework with a controlled experiment to assess the efficacy of the methods. To interpret the experiment results, we utilize one-way repeated measures analysis of variance. Also, a questionnaire survey with experts was conducted to evaluate the usability of the methods.