Visualizing Time-Dependent Data Using Dynamic t-SNE

Визуализация зависящих от времени данных с использованием динамического t-SNE
Alexandre X. Falcão, Alexandru Cristian Telea, Paulo Rauber
2016-01-01

dimensionality reductiondynamic t-SNEprojection reliabilitytemporal coherencetime-dependent datasets
Many interesting processes can be represented as time-dependent datasets. We define a time-dependent dataset as a sequence of datasets captured at particular time steps. In such a sequence, each dataset is composed of observations (high-dimensional real vectors), and each observation has a corresponding observation across time steps. Dimensionality reduction provides a scalable alternative to create visualizations (projections) that enable insight into the structure of such datasets. However, applying dimensionality reduction independently for each dataset in a sequence may introduce unnecessary variability in the resulting sequence of projections, which makes tracking the evolution of the data significantly more challenging. We show that this issue affects t-SNE, a widely used dimensionality reduction technique. In this context, we propose dynamic t-SNE, an adaptation of t-SNE that introduces a controllable trade-off between temporal coherence and projection reliability. Our evaluation in two time-dependent datasets shows that dynamic t-SNE eliminates unnecessary temporal variability and encourages smooth changes between projections.
1
Dynamic t-SNE adapts t-SNE for time-dependent datasets by controlling the trade-off between temporal coherence and projection reliability.
2
Evaluation on two time-dependent datasets shows that dynamic t-SNE eliminates unnecessary temporal variability and promotes smooth changes between projections.
3
Independent t-SNE projections across time can introduce unnecessary variability, making data evolution difficult to track.

time-dependent datasets, represented as sequences of corresponding high-dimensional observations across time steps

the temporal coherence, projection reliability, and unnecessary temporal variability of dimensionality-reduction visualizations across successive time steps

Publication Details
Publication Date
2016-01-01
Journal
Publisher
ISSN
Cited by
138
Access Type
Author Information
Authors
Alexandre X. Falcão
Alexandru Cristian Telea
Paulo Rauber
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%