Networks are graphs consisting of vertices which represent objects and edges representing the relationship between these objects. As more data is being made available in all fields of sciences and engineering,networks are being constructed using the data and network analysis approaches have been developed to analyze these networks as they reveal interesting things that are not seen before. In digital humanities, massive historical data is being used to construct networks and answer interesting questions. In this blog we describe three such cases, where network analysis is emerging as a tool in digital humanities.
First we describe network analysis in Venetian maritime networks. In this paper the authors focused on the period starting from the end of 13th century and the fall of Constantinople in 1453. The Venetian Republic used an auction system to allocate space on the ships running in the shipping lanes linking Venice and west Mediterranean regions, England and Flanders. The data obtained from the venetian state archives originally, was transformed into structured data by using computer vision techniques and Optical Character Recognition software. Important quantities like year, number of ships, stopping points of the ships and duration at the stopping points were extracted and cleaned for unwanted data points. Now, the authors paired consecutive stopovers to build a network of all the ships during this 170 year period(shown in the figure below). The Vertices of the network are the ports at which the ships stopped during this period and the edges representing the traffic between the connected nodes.
From this global network, sub-networks were obtained for each year of navigation. Now the traffic at particular stops(Crete and Cyprus) was analyzed in relation to historical events. The authors also evaluated a measure to represent the commercial betweenness of these two islands and evaluated them for each year during this 170 year period. The reasons for increase or decrease of traffic in these two ports was attributed to historical events like opening of alternative routes, and use of these ports as hubs towards different routes during specific years in the maritime history of venice.
The second paper describes the SeNeReKo project which is developing methods for the application of network analysis to study religious texts. Formation and development of religious traditions were significantly influenced by inter-religious contacts. Network analysis can be used as a powerful tool to analyze such interactions between religions. In this project computational linguistic methods are applied to identify the vertices and edges of the network. Semantics of the words are embedded in the network. Networks of words are created after annotating the religious texts of the target language. The analysis involves centrality measures and clustering algorithms to discover semantic structures.
A few selected ancient Egyptian texts and The Buddhist Pali Canon are used as two independent test cases to test this approach. These texts are converted to TEI compliant XML to have a generic starting point of analysis. The Canon contains information about the comparisons between Buddha and other religious groups in Northern India and Sri Lanka. In the Egyptian case it is used to analyze the religious dynamics in the history of Ancient Egypt. This project develops new analytical tools for network analysis in digital humanities which can be used at other places unlike the project modelling maritime network. This project aims at giving new insights into influence of one religion on the other.
“One of the most potent ideas in the social sciences is the notion that individuals are embedded in thick webs of social relations and interactions” —–Borgatti et al 
Now we describe social network analysis from the third paper . Social network analysis is increasingly found in historical research since it has explained how Cosimo de Medici took advantage of the gaps in the Medicis’ social network to take political control of Florence. The third paper deals social networks of people of medieval Scotland. In this paper the authors constructed a social network based on the patterns of people visiting the same charter during the period 1093 and 1314. From this analysis it was found that Duncan, Earl of Fife has much bigger social network than other important people, raising questions about the possible roles he played in the Scottish government during William I’s reign.
The common theme across these three papers is summarized below.
- Obtain the data from literature using Computer-vision, Character recognition techniques and convert historical data into digital formats following certain standards.
- Identify the vertices and edges of the network.
- Analyze the network of various features or perform operations on the network to find answers for the questions you want to answer.
- A Network Analysis Approach of the Venetian Incanto System (http://dharchive.org/paper/DH2014/Paper-424.xml)
- Network Analysis for the History of Religions -The SeNeReKo Project (http://dharchive.org/paper/DH2014/Poster-226.xml)
- Borgatti, S. P. and A. Mehra, D.J. Brass and G. Labianca (2009) “Network Analysis in the Social Sciences”, Science, Volume 323, pp. 892 – 895
- Using Social Network Analysis to Reveal Unseen Relationships in Medieval Scotland (http://dharchive.org/paper/DH2014/Paper-881.xml)