Master Thesis Defence: "Conceptualization and Evaluation of Idea Similarities based on Semantic Enrichment & Knowledge Graphs" by Maximilian Stauss (Mar 10, 2021)
News from Mar 03, 2021
March 10th, 11 am, the Master Thesis Defence: "Conceptualization and Evaluation of Idea Similarities based on Semantic Enrichment & Knowledge Graphs" by Maximilian Stauss will take place at HCC.lab.
Abstract
This master thesis gives an overview of different semantic measures on general knowledge graphs like Wikidata. The focus was on the comparison of short idea texts. For these pairs of ideas, semantic measures were implemented which should represent human intuition as well as possible. To test the quality of the implemented measures, they were compared with a state-of-the-art algorithm from the field of machine learning on the one hand and with the results of a user study on the other hand. The basis of this work is the book ‘Semantic Similarity from Natural Language and Ontology Analysis’. [20], which provides an overview of different semantic measures on Knowledge Graphs. By using this approach to compare short idea texts, the author hopes to show alternatives to commonly used machine learning approaches.
[20] Sébastien Harispe et al. “Semantic similarity from natural language and ontology analysis”. In: Synthesis Lectures on Human Language Technologies 8.1 (2015), pp. 1–254.
First assessor: Prof. Dr. Claudia Müller-Birn
Second assessor: Prof. Dr. Marian Margraf
Location: WebEx (If you want to join the thesis defence, please contact Phil Wernberger.)