WordBridge: Using Composite Tag Clouds in Node-Link Diagrams for Visualizing Content and Relations in Text Corpora

Kim, Kyung-Tae; Ko, Sungahn; Elmqvist, Niklas; Ebert, David S.

doi:10.1109/hicss.2011.499

Cited by 16 publications

(5 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides the automatic layout customization, further options to customize, enrich, and (hopefully) improve word cloud representations could include the use of user-driven customizations, 23,24 typographic attributes, 25 word-scale/sparkline representations, 26,27 or even node-link representations, such as word webs by Baumer et al 28 Finally, besides the techniques focusing on the word cloud as the final result in itself, the prior work in information visualization and visual analytics has proposed various approaches for the actual use of word clouds for analytical tasks (mainly for text analysis). For example, WordBridges by Kim et al 29 used word clouds to represent details for particular network nodes and edges within a node-link diagram: here, a co-authorship network for several researchers could be represented while demonstrating word clouds based on the publication keywords for an individual researcher (nodes) and co-authored publications (edges). Besides bibliographic data, applications of this approach were demonstrated for intelligence reports and works of fiction.…”

Section: Advanced Techniques and Applicationsmentioning

confidence: 99%

From word clouds to Word Rain: Revisiting the classic word cloud to visualize climate change texts

Skeppstedt,

Ahltorp,

Kucher

et al. 2024

Information Visualization

View full text Add to dashboard Cite

Word Rain is a development of the classic word cloud. It addresses some of the limitations of word clouds, in particular the lack of a semantically motivated positioning of the words, and the use of font size as a sole indicator of word prominence. Word Rain uses the semantic information encoded in a distributional semantics-based language model – reduced into one dimension – to position the words along the x-axis. Thereby, the horizontal positioning of the words reflects semantic similarity. Font size is still used to signal word prominence, but this signal is supplemented with a bar chart, as well as with the position of the words on the y-axis. We exemplify the use of Word Rain by three concrete visualization tasks, applied on different real-world texts and document collections on climate change. In these case studies, word2vec models, reduced to one dimension with t-SNE, are used to encode semantic similarity, and TF-IDF is used for measuring word prominence. We evaluate the technique further by carrying out domain expert reviews.

show abstract

Section: Advanced Techniques and Applicationsmentioning

confidence: 99%

From word clouds to Word Rain: Revisiting the classic word cloud to visualize climate change texts

Skeppstedt,

Ahltorp,

Kucher

et al. 2024

Information Visualization

View full text Add to dashboard Cite

show abstract

“…They may show additional selection‐related nodes and edges that are not visible in the global network, or display quantitative aspects of sub‐edges. Semantic context can, for example, provide the terms used by two actors in their correspondences. In cases where edges are derived from textual corpora, it may be useful for validation to show the sentence from which a relation between two entities was extracted from, e.g., in form of word clouds [KKEE11]. Multivariate context, including overlaps and blends of temporal, geospatial, structural, and semantic context, can be visualized through individually crafted embedded visualizations, using multiple encoding strategies, for example, parallel coordinates offering an overview of multivariate attributes.…”

Section: Opening the Design Space For Unfolding Graph Edgesmentioning

confidence: 99%

“…In terms of graph visualizations, hybrid techniques [DGDMT21], such as NodeTrix [HFM07] In contrast to enhancement techniques for nodes, there is still a scarcity of comparable techniques to reveal multivariate edge attributes [NSML19]. However, there is one promising example that is quite similar to our idea of unfolding edges: WordBridge dynamically expands edges in NLD into word clouds to add textual context information [KKEE11]. This example indicates that there is potential in dynamically adding context to edges, however, the broader design space of such approaches still remains largely unexplored.…”

Section: Embedding Additional Informationmentioning

confidence: 99%

Unfolding Edges: Adding Context to Edges in Multivariate Graph Visualization

Bludau

Dörk

Tominski

2023

Computer Graphics Forum

View full text Add to dashboard Cite

Existing work on visualizing multivariate graphs is primarily concerned with representing the attributes of nodes. Even though edges are the constitutive elements of networks, there have been only few attempts to visualize attributes of edges. In this work, we focus on the critical importance of edge attributes for interpreting network visualizations and building trust in the underlying data. We propose ‘unfolding of edges’ as an interactive approach to integrate multivariate edge attributes dynamically into existing node‐link diagrams. Unfolding edges is an in‐situ approach that gradually transforms basic links into detailed representations of the associated edge attributes. This approach extends focus+context, semantic zoom, and animated transitions for network visualizations to accommodate edge details on‐demand without cluttering the overall graph layout. We explore the design space for the unfolding of edges, which covers aspects of making space for the unfolding, of actually representing the edge context, and of navigating between edges. To demonstrate the utility of our approach, we present two case studies in the context of historical network analysis and computational social science. For these, web‐based prototypes were implemented based on which we conducted interviews with domain experts. The experts' feedback suggests that the proposed unfolding of edges is a useful tool for exploring rich edge information of multivariate graphs.

show abstract

“…In these cases, the temporal evolution(s) of the data points are plotted along an axis to show the dynamics. Alternatively, text and topic data can be plotted via ordination or network-based plots [1,10,18,21,31,58]. To accommodate streaming data, the visualization must update using animation or a display refresh.…”

Section: Visualizing Text and Meeting Datamentioning

confidence: 99%

TalkTraces

Chandrasegaran

Bryan

Shidara

et al. 2019

Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems

View full text Add to dashboard Cite

Topic (columns) vs. Agenda (rows) scatterplot of discussion Transcript of current and previous session Transcript-to-agenda relatedness for previous meeting Real-time transcript-to-agenda relatedness for current meeting transcript relatedness for current relatedness for Previous meeting agenda Current meeting agenda Word cloud for previous meeting Topic (columns) vs. Agenda (rows) scatterplot of discussion 1 2 3 Local/global topic keyword control slider Relevant keywords for topics Real-time speech-totext transcription text transcription Transcript of current and previous session Real Real Local/global topic Relevant keywords for topics Word cloud for current meeting Figure 1: The final iteration of TalkTraces shows (1) agenda items from the previous and current meeting, (2) topics computed from the previous meeting transcript, and (3) transcript lines rendered as pie charts reflecting topic distribution, aligned with the most related agenda item. Streaming audio transcribed from the current meeting updates the visualizations in real time.

show abstract

WordBridge: Using Composite Tag Clouds in Node-Link Diagrams for Visualizing Content and Relations in Text Corpora

Cited by 16 publications

References 23 publications

From word clouds to Word Rain: Revisiting the classic word cloud to visualize climate change texts

From word clouds to Word Rain: Revisiting the classic word cloud to visualize climate change texts

Unfolding Edges: Adding Context to Edges in Multivariate Graph Visualization

TalkTraces

Contact Info

Product

Resources

About