Title: Clustering Graphs
Subtitle: Applying a Label Propagation Algorithm to Detect Communities in Graph Databases
The thesis is on the detection of collaboration communities in the academia by making use of graph databases such as Neo4j or ArangoDB and a Label Propagation Community Detection Algorithm. To visualize the results, a full webapp was built made of a frontend interface with React, TypeScript and CytoscapeJS consuming a custom built GraphQL API with NodeJS and Express.
Pdf file of the thesis: 2021-09-14_masters_thesis_document.pdf
LaTeX source code of the thesis: ./masters_thesis
Presentation slides (only): 2021-09-23_thesis_presentation_slides_only
Presentation slides (with notes): 2021-09-23_thesis_presentation_slides_and_notes.pdf
Presentation slides (with notes x4 per page): 2021-09-23_thesis_presentation_slides_and_notes_x4_per_A4_page.pdf
LaTeX source code of the presentation slides: ./thesis_presentation
Preliminary data conversion (XML to JSON): ./convert_large_xml_to_json
Dataset import in ArangoDB: ./arangodb_import_json_data
Data manipulations towards obtaining the graph: ./distribute_data_in_arangodb
Source code of the GraphQL API back-end: ./adomainthat-rocks_backend
Source code of the front-end interface: ./adomainthat-rocks_frontend
in detail
in production
zoomed in graph of collaborations
Note: This thesis was awarded with the Rovelli Award as the best graduation thesis of Computer Science and Engineering Degree of 2021 related to topics of software development. Prize was awarded by the University of Bergamo and WebResults.