This document discusses character network analysis of Émile Zola's novel series Les Rougon-Macquart. It explains that character networks can be created by representing characters as nodes and their proximity in the text as edges. This creates a social network of characters. It also discusses applying measures like betweenness centralization and coreness to character networks to analyze novels and classify them based on character interactions. For Les Rougon-Macquart specifically, the document outlines the process used to create the character network from the text, including optical character recognition and cleaning. It analyzes the network using betweenness centralization and coreness and finds they highlight different properties of Zola's narration, showing diversity within the series.
OKFN Greece meet-up
Friday, April 6, 2012, 5:00 PM
Aristotle University of Thessaloniki, Research Dissemination Center
Prof. I. Antoniou (Director of MSc Web Science, AUTH, Steering Committee OKFN Greece). The power of Openness. Open Data and Open Knowledge
OCR is abbreviated as Optical Character Recognition. Optical Character recognition is a process of recognition of different characters (printed or handwritten) from a digital image of documents. In OCR technique, characters can be recognized through optical mechanism. Various combinations of lines & curves make the characters. Characters recognition ability of human beings is very high. They can recognize all characters accurately. But same task is very difficult by OCR system. The wide usage of touch-screen based mobile devices has led to a large volume of the users preferring touch-based interaction with the machine, as opposed to traditional input via keyboards/mice. To exploit this, we focus on the Android platform to design a personalized handwriting recognition system that is acceptably fast, light-weight, possessing a user-friendly interface with minimally-intrusive correction and auto-personalization mechanisms.
OKFN Greece meet-up
Friday, April 6, 2012, 5:00 PM
Aristotle University of Thessaloniki, Research Dissemination Center
Prof. I. Antoniou (Director of MSc Web Science, AUTH, Steering Committee OKFN Greece). The power of Openness. Open Data and Open Knowledge
OCR is abbreviated as Optical Character Recognition. Optical Character recognition is a process of recognition of different characters (printed or handwritten) from a digital image of documents. In OCR technique, characters can be recognized through optical mechanism. Various combinations of lines & curves make the characters. Characters recognition ability of human beings is very high. They can recognize all characters accurately. But same task is very difficult by OCR system. The wide usage of touch-screen based mobile devices has led to a large volume of the users preferring touch-based interaction with the machine, as opposed to traditional input via keyboards/mice. To exploit this, we focus on the Android platform to design a personalized handwriting recognition system that is acceptably fast, light-weight, possessing a user-friendly interface with minimally-intrusive correction and auto-personalization mechanisms.
2. ● Nodes symbolize characters
● Edges symbolize proximity in the discourse
=> social network of characters
Character networks
3. ● Most novels can be turned into networks
● Different orders, different sizes
● A network provides a signature of the novel
● Application of measures
Character networks
19. “The k-core of a graph is a
maximal subgraph in which
each vertex has at least
degree k.
The coreness of a vertex is k
if it belongs to the k-core but
not to the (k+1)-core.” (Csardi,
2006)
Coreness
22. Classification through main protagonist and
cores of protagonists.
On this corpus, coreness and betweenness
centralisation highlight different properties.
Diversity within Zola’s narration.
Conclusion
23. Take into account weights of
edges for a better accurracy of
“coreness”.
How do we study interlocking
novels ? (Zola, Balzac, …)
Representation and study of
temporality.
Future works