Browsing by master's degree program "Master's Programme Linguistic Diversity in the Digital Age"

Graph databases are an emerging technology enticing more and more software architects every day. The possibilities they offer to concretize data is incomparable to what other databases can do. They have proven their efficiency in certain domains such as social network architecture where relational data can be structured in a way that reflects reality better than what Relational Databases could provide. Their usage in linguistics has however been very limited, nearly inexistent, regardless of the countless times where linguists could make great use of a graph. This paper aims to demonstrate some of the use cases where graph databases could be of help to computational linguistics. For all these reasons, this thesis focuses on practical experiments where a Graph Database (in this case, Neo4j) is used to test its capabilities to serve linguistic data. The aim was to give a general starting point for further research on the topic. Two experiments are conducted, one with a continuous flow of relational textual data and one with a static corpus data based on the Universal Dependencies Treebanks. Queries are then performed against the database and the retrieval performances are evaluated. User-friendliness of the tools are also taken into account for the evaluation.