Tags

, , , ,

Every passing day, the importance of data visualization is increasing more and more. The digitization of data and storing the data in large amounts necessitate to devise effective and efficient methods to analyze large amount of data. One of the promising class of methods are the visualization methods, which can convert large amount of data to visual representations so that user can analyze the large amount of data more easily. In the rest of this blog post, firstly, the visualization methods which are used in the area of digital humanities will be presented, then two cases, one is a real case: economic crisis and the other one is fictional case: emotions of London, in which the visualization techniques are used will be discussed.

In the abstract [1], the various visualization methods and interaction methods are discussed. Also, the evaluation methods for the performance of visualization methods, used in the area of digital humanities, are presented. Histogram, word cloud, scatter plot, timeline, geographic map, 3D model, parallel coordinates, infographic, tree, matrix, heatmap, graph, sankey diagram, cluster map and animation are some of the visualization techniques. When it is looked at the digital humanities research, it is seen that geographic map, 3D model and graph are the most widely used methods. This is simply because graphs are one of the best ways to represent the social network between people. Geographic map and 3D model are two of the most suitable ways of modelling the places where people live. Also, it should be noted that the methods which are used in the abstracts [2] and [3] are graph and geographic map, respectively. Interaction methods can be summarized as abstract/elaborate, explore, filter, encode, reconfigure, connect and select/focus. Abstract/eloborate -which is defined as when selecting an object, show less or more details about object- and explore -which can be stated as show the details about chosen part of the map- are highly used methods in the digital humanities research. In the evaluation part, the prototypes of visualization methods are assessed by researchers with using these prototypes in real research questions. Although, it was seen that only few of prototypes works well, since visualization methods make the path smooth in research, it can be stated that the visualization methods will be promising tool in the digital humanities research.

In the abstract [2], it is discussed about a practical example of generating graph for the 2007-2008 economic crisis to find who or what caused the economic crisis. By using large amount of documents about economic crisis and Standford Named Entity Recognizer, entities are generated. Then, graph is generated from these entities by using Gephi software. In the graph, nodes are used the demonstrate group of people and corporations, and edges between two nodes are used to show the number of co-occurences of two nodes (person or corporation) in the same sentence in the documents. The most important advantage of graph was that in addition to showing obvious links such as the relation between chief executive officer and his/her corporation, which was expected, the graph achieved to show the relation between some people and organizations, which was difficult to find. Moreover, researchers could estimate the beginning time of financial crisis by looking at temporal visualization of data. In the abstract [1], it is stated that one of the most widely used visualization method is graph, and in the abstract [2], one of the implementation of graph is presented.

In the abstract [3], the geographic map, one of the most widely used visualization techniques as stated in the abstract [1], is used. The geographic map of places in London which are stated in literature texts is formed by using the literature texts from years of 1700 to 1900. In addition, the emotion mapping of London is generated from fictional literature texts. In order to generate geographic map, a list of places are chosen and by using Standford Named Entity Recognizer, which is also used in the abstract [2], entities are formed, then geographical map is generated from these entities. In the map, circles are used to show discrete places such as buildings, streets etc., and polygons are used to demonstrate wide areas such as districts etc. While the source documents for the visualization in the abstract [2] come from the real situation of financial crisis, the source documents for the visualization in the abstract [3] are literature documents which tell fictional situations. Also, by using literature texts of 1700-1900, the geographic map of emotions in London is produced. In this map, there are only polygons, and colors of green, red and gray which demonstrate happiness, fear and neutral. It is found that the places of prisons, hills etc. are linked with fear and the places of parks, squares, theatres etc. are linked with happiness. Consequently, the researchers could use these maps to alleviate the difficulties in analyzing the literature texts so it can be stated that the visualization techniques can also be useful in the digital humanities research.

In conclusion, there exists many visualization methods. Graph, 3D model and geographic map are the most widely used visualization techniques in the digital humanities research. Since the visualization methods help researchers to analyze large amount of data more easily and the data in the digital humanities research is in large amount, it can be stated that the visualization methods will gain more and more attention in the digital humanities research.

REFERENCES

[1] K. Verbert, “On the Use of Visualization for the Digital Humanities”, 2015. [Online]. Available: http://dh2015.org/abstracts/xml/VERBERT_Katrien_On_the_Use_of_Visualization_for_t/VERBERT_Katrien_On_the_Use_of_Visualization_for_the_Dig.html

[2] T. Poibeau and P. Ruiz, “Generating Navigable Semantic Maps from Social Sciences Corpora”, 2015. [Online]. Available: http://dh2015.org/abstracts/xml/POIBEAU_Thierry_Generating_Navigable_Semantic_Maps_from_Social_Sciences_Corpora/POIBEAU_Thierry_Generating_Navigable_Semantic_Maps_from_Social_Sciences_Corpora.html

[3] R. Heuser, M. Algee-Hewitt, V. Tran, A. Lockhart and E. Steiner, “Mapping the Emotions of London in Fiction, 1700-1900: A Crowdsourcing Experiment”, 2015. [Online]. Available: http://dh2015.org/abstracts/xml/HEUSER_Ryan_James_Mapping_the_Emotions_of_London_/HEUSER_Ryan_James_Mapping_the_Emotions_of_London_in_Fic.html

Advertisements