Storing actionable metrics in a graph database
Malmström, Sonja (2020)
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
This thesis was done in collaboration with the Norwegian company Limber AS, developer of the cloud-based web application Limber Projects. This web application can be used by companies who work with complex and supplier dependent projects. Limber was interested in future possibilities to study and evaluate user behaviour in their web application by storing relevant user data in a graph database. A graph database is a type of non-relational database that mainly focuses on storing information about relation-ships between data objects as nodes and edges. The goal with this thesis was to provide an overview of how user behaviour data can be stored in a graph database and how the data modeling process can be done. During the data modeling process of this thesis it was decided what data would be stored as nodes and edges in order to provide infor-mation of how users move in the web application during a session. The practical part of this thesis consisted of creating a prototype of a working graph database with the purpose of storing user behaviour data. This prototype was developed in ArangoDB and test data created by the author was used to evaluate the setup of the data model as well as test the performance of the database. This prototype made possible an understanding of what data is important enough to become its own data object.