who are the leading CHARACTERs in Les miserables?


Lab Reports, Networks

Introduction

Victor Hugo’s purpose in writing Les Misérables is to expose and critique the social injustice that existed in nineteenth-century European society, with a focus on the sufferings of the poor. Although the book is set in France, Hugo said it applied to all of Europe. My attempt for this dataset manipulation is simple. I aim to let those people who have not seen or read Les Miserables would be able to quickly grasp the tension between each character and who the main characters are in this historical novel. 

Visualization References

Figure.1 Gephi Fruchterman-Reingold layout based on initial Geolayout

Figure.2 Gephi Force Atlas 2 Large Steam network with Google map

I came across these two graphs while searching for my visual references on the website. These two immediately captivated my attention perhaps the color richness and the clear communication of community divisions. The left graph(Figure.1) node betweenness centrality mapped to circle size; closeness centrality of nodes/edges mapped to circle/line color (Rheinwalt et al., 2012). The graph(Figure.2) uses Google Maps API and Gephi Force Atlas 2 to visualize a relatively large network collected from Steam Community members.

Process and Visualization Methodology

The Dataset

The dataset I used for this report is from here. This undirected network contains 74 nodes and 248 edges, which represents co-occurrences of characters in Victor Hugo’s novel ‘Les Misérables’. A node means a character and an edge between two nodes shows that these two characters appeared in the same chapter of the book. The weight of each link indicates how often such a co-appearance occured. The dataset also sorts gender. That’s how I used to differentiate the gender by color difference. The original dataset is a GML file which allows me to import to Gephi directly without struggling with R and OpenRefine.

NameLes Misérables
Network formatUnipartite, undirected
Edge typePositive weights, no multiple edges
Size n = 74
Volumem = 248
Loop countl = 0
Diameterδ = 5

Tool

Gephi is an interactive tool used to explore, analyse, spatialize, filter, and export all types of network. It is great if you have collected data that examines the connections between points that make up a network, such as informational, biological or social networks. To better use this tool it is helpful to understand the basics of network graphs such as the differences between “nodes” and “edges”. A node is a point, or a location, like a source or a target, and edges are connections between nodes. The higher the frequency of connections from each node the more compact your graph will look and the harder it will be to see what’s going on.  Gephi offers users options to change the design of the graph to increase clarity and readability. Users can change the design of nodes, edges and labels through changing the colors, size and fonts.

Choice of Color  

My inspiration of color scheme is from these two graphs I searched via pinterest and colorpalette.cinema. I think the colors well address the plot and the background of this historical novel- the working class suffered from poverty and injustice in 19th France. From the samples, you can see the color are leaning toward faded, cold and gloomy tone. The tarnish palette on Gephi represents appropriately, I used it to generate the color for my graphs.

Results

  • Force Atlas 2 – Gender
  • Fruchterman-Reingold – Modularity
  • Fruchterman-Reingold – Network Diameter

The major characters in this novel are Valijean, Javert, Fantine, Cosette, Marius, Gavroche and Bishop Myriel. Interestly, the spatial graphs did reveal this part.  The weightiness of the edges reflect the tension between characters appearing in the same chapter of the book. The first graph, I chose blue and red combination with white background which stands for the national flag of France. I ran Label Adjust and Noverlap on each graphs to avoid cluster. However, the label of first graph still meshed up even I manually adjusted the nodes distance and with Arial Narrow font. In order to show the clusters dynamics I use Force Atlas 2 to begin with.

Reflection

The original dataset I got is about Chinese Buddhism history spanning from different dynasties of ancient China. But, the dataset contains too many nodes and edges. My laptop doesn’t hold up enough memory and computing power to run the dataset so I yielded. Despite how much I want to use Gephi to create marvelous spatial graphs. I decided to use a small dataset to focus on exploring the features, observing, and applying different statistics filters to acquaint myself with this new software. Personally, I think the interface design of Gephi is not very user-friendly and dated. It has a lot of features to offer but novice users require a learning curve. There are jargons that overwhelm you, such as Fruchterman Reingold, what is this? I suspect it is a human name, maybe named after someone who invented this spatializing pattern. I increased my readability and changed my layout from Force Atlas 2 to Yifan Hu Proportional after peers usability testing review. Yifan Hu Proportional well-interpreted the dataset more clear.

Sources

GEPHI – introduction to network analysis and visualization. Martin Grandjean. (n.d.). Retrieved November 9, 2021, from http://www.martingrandjean.ch/gephi-introduction/.

Gephi. Visualising Information for Advocacy. (n.d.). Retrieved November 9, 2021, from https://visualisingadvocacy.org/node/615.html.

Layouts with different types of forces. layouts with … (n.d.). Retrieved November 10, 2021, from https://researchgate.net/figure/Layouts-with-different-types-of-forces-Layouts-with-Fruchterman-Reingold-a-r-3_fig1_262977655.

Networks. (n.d.). Retrieved November 9, 2021, from http://konect.cc/networks/.

Node-link diagram for a German measurement stationbased … (n.d.). Retrieved November 11, 2021, from https://researchgate.net/figure/Node-link-diagram-for-a-German-measurement-stationbased-precipitation-network-Gephi_fig4_276880365.

PAF 2021 – PROJ102. Force Atlas: Graph layout algorithm | PAF 2021 – PROJ102. (n.d.). Retrieved November 10, 2021, from https://paf.telecom-paris.fr/projets/force-atlas-graph-layout-algorithm.

Wikimedia Foundation. (2021, November 9). Les Misérables (musical). Wikipedia. Retrieved November 11, 2021, from https://en.wikipedia.org/wiki/Les_Mis%C3%A9rables_(musical).

Wikimedia Foundation. (2021, October 23). Les Misérables. Wikipedia. Retrieved November 11, 2021, from https://en.wikipedia.org/wiki/Les_Mis%C3%A9rables.

Wright, O. A. J., Anise, O., Wright, J., Authors Olabode Anise Data Scientist  @justsayo Jordan Wright Principal R&D Engineer , Olabode Anise Data Scientist  @justsayo Jordan Wright Principal R&D Engineer  @jw_sec, Olabode Anise Data Scientist  @justsayo, & Jordan Wright Principal R&D Engineer  @jw_sec. (2018, September 13). Mapping social networks with Gephi. Duo Security. Retrieved November 9, 2021, from https://duo.com/blog/mapping-social-networks-with-gephi.