Real-Time Visualization of Streaming Text with a Force-Based Dynamic System

Abstract

An interactive visualization system, STREAMIT, enables users to explore text streams on-the-fly without prior knowledge of the data. It incorporates incoming documents from a continuous source into existing visualization context with automatic grouping and separation based on document similarities. STREAMIT supports interactive exploration with good scalability: First, keyword importance is adjustable on-the-fly for preferred clustering effects from varying interests. Second, topic modeling is used to represent the documents with higher level semantic meanings. Third, document clusters are generated to promote better understanding. The system performance is optimized to achieve instantaneous animated visualization even for a very large data collection. STREAMIT provides a powerful user interface for in-depth data analysis. Case studies are presented to demonstrate the effectiveness of STREAMIT.

Visualizing Hidden Themes of Taxi Movement with Semantic Transformation

“…Consequently, each keyword is assigned a probability of its appearance in each topic. Visualizing the topics has been investigated to present the themes inside news, tweets, and other textual documents [6,1]. To the best of our knowledge, we are the first to apply this powerful tool to geographic trajectory data, enabled by the semantic transformation.…”

A visual analytics approach to exploring protein flexibility subspaces

“…Alsakran et al [1] use the algorithm to visualize streaming text documents where the closeness of node locations indicates similarity. Their system uses a force-based physical simulation of particles that represent data items.…”

“…In [1] the visualization is updated after each text document particle enters the system at different time steps. The data set used here is not time dependent but the application is similar because the node positions are evolved from one layout to another.…”

Visualização da evolução temporal de coleções de artigos científicos

“…Buscando uma solução incremental que permitisse a adição e remoção de elementos durante a geração da visualização, Alsakran et al (2012) desenvolveram a ferramenta Streamit para visualizar coleções de documentos em streaming, que constantemente recebem novos documentos.…”

“…Apesar da componente temporal ter um papel crucial em muitos tipos de dados, inclusive em textos, muitas técnicas de projeções multidimensionais existentes não a consideram explicitamente em seu processo de mapeamento. Existem algumas poucas técnicas de projeção multidimensional dinâmicas (Alsakran et al, 2012;Pinho et al, 2010;Joia et al, 2011;Paulovich et al, 2010;Leydesdorff;Schank, 2008;Albrecht-buehler et al, 2005) e todas elas buscam obedecer duas propriedades principais conflitantes:…”