Using social co-occurrence networks to analyze Biblical narrative
|has title::Creating Using social co-occurrence networks to analyze Biblical narrative|
|Master:||project within::Computational Intelligence and Selforganisation|
|Student name:||student name::Frederik de Vree|
The aim of this study is to see if the genre of narrative books can be better understood by creating social co-occurrence networks from them: social networks where the nodes are names occurring in the text, and the strength of connections between the nodes are defined by their proximity in the text by amount of sentences.
To test the relevance of using social co-occurrence networks, four uses were examined: determining the importance of characters in a narrative, finding the start of a new narrative, finding parallel narratives and recognizing communities. Theological literature provided test cases and guidelines on what to look for and where. Literature on network theory presented methods to construct the social co-occurrence networks and to analyze them in such a way that the cases could be tested.
The results show the presented method can be used to find the most likely candidates for important characters, but it is not reliable enough to always find the single correct one. Social co-occurrence networks do not help with finding the start of a new narrative, because there are too much differences inside a story to notice the transition. The networks can however be used to correctly identify parallel stories in most cases. Finally, genealogies can reliably be identified in multiple ways, proving that it is possible to recognize communities.
Social co-occurrence networks cannot replace existing, more reliable and detailed, but more labor-intensive approaches. It does however provide an additional tool to get an overview of narrative Bible books by providing the important characters and communities in texts and parallels between texts.