NetworkNarratives: Data Tours for Visual Network Exploration and Analysis (NetworkNarratives: Data Tours for Visual Network Exploration and Analysis)

Original link: http://vis.pku.edu.cn/blog/networknarratives%EF%BC%9A%E8%A7%86%E8%A7%89%E7%BD%91%E7%BB%9C%E6 %8E%A2%E7%B4%A2%E4%B8%8E%E5%88%86%E6%9E%90%E7%9A%84%E6%95%B0%E6%8D%AE%E4%B9 %8B%E6%97%85networknarratives-data-tours-for-visual-networ/

Exploring the network takes a lot of effort, and based on strong expertise, such a process is time-consuming and challenging. In order to solve these problems, the author developed a semi-automatic data cruising system NetworkNarratives[1,2] to help network analysts explore complex networks. This paper [1] is the first to propose data tours in the field of network analysis.

The NetworkNarratives system is oriented towards tasks such as network overview, self-network analysis, and node cluster exploration, allowing network analysts to browse interactive slideshows containing network factual information using visualizations and text annotations. On each slide, network analysts are free to explore the network and designate nodes, links, or subgraphs as seed elements for subsequent tours.

Figure 1 shows an overview of the interface of the system. Users first select and import the datasets they are interested in in the dataset selection module (a), and then select the data exploration routes they care about in the cruise route selection module (b). The exploration details of the route will be displayed in the cruise overview module (c), and at this point, the user can start to explore the network data. The interactive visualization module (d) and the text description module (e) are charts and text descriptions of the data information during the parade, respectively. Users can click the button in the navigation module (f) to switch slides.

Figure 1: System interface of NetworkNarratives

What does the specific cruising route of network data include? The paper systematically summarizes 10 data exploration routes for network data, as shown in Figure 2.

Figure 2: 10 Network Data Parade Routes

Taking the three data cruising routes as examples, Figure 3 shows the description of using the NetworkNarratives system for data cruising. Starting from the red line, start the journey of network overview exploration (A), each slide has its title and text description, and the number in the circle represents the slide number in the exploration journey. When focusing on a specific node or network subgraph, the user can turn to the cruise route related to the node or subgraph for browsing (B, C). In addition, if the user wants to further dig into the detailed information in the network data, he can browse the additional slides (10.1, 10.2) of the parade route on demand by means of “detour”.

Figure 3: An example of data patrolling using the NetworkNarratives system

The authors then evaluate the NetworkNarratives system for network analysis experts and network analysis novices. Compared with free-form, data tours (tours) generally received higher positive subjective evaluations, such as positioning and narrative, stimulating further exploration and discovery, or helping learning analysis.

Figure 4: Advantages and disadvantages of data tours (tours) and free-form exploration (free-form) in user experiments

Overall, this article takes data tours as a supplement to recommendation methods and free-form exploration in the form of semi-automated data tours, helping to explore and analyze network data, and creates a new era of data tour exploration in the context of network data structures. First of its kind. Furthermore, the system strikes a balance between statically defined tours, data-driven recommender systems, and open-ended free-form exploration by incorporating semi-automated techniques to retrieve relevant facts based on user needs and allow basic forms of personalization.

references:

[1] Wenchao Li, Sarah Schöttler, James Scott-Brown, Yun Wang, Siming Chen, Huamin Qu, and Benjamin Bach. 2023. Network Narratives: Data Tours for Visual Network Exploration and Analysis. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (CHI ’23), April 23–28, 2023, Hamburg, Germany.

[2] https://networknarratives.github.io/

This article is reproduced from: http://vis.pku.edu.cn/blog/networknarratives%EF%BC%9A%E8%A7%86%E8%A7%89%E7%BD%91%E7%BB%9C%E6 %8E%A2%E7%B4%A2%E4%B8%8E%E5%88%86%E6%9E%90%E7%9A%84%E6%95%B0%E6%8D%AE%E4%B9 %8B%E6%97%85networknarratives-data-tours-for-visual-networ/
This site is only for collection, and the copyright belongs to the original author.