- 49 views
- 87 downloads
Knowledge Graphs: Construction and Applications
- Author / Creator
Knowledge graphs become a de facto standard for representing data in situations where relations between individual pieces of information are important. These semantically rich data structures provide opportunities to develop methods and algorithms for new ways of analyzing and utilizing data.
In this work, we illustrate benefits of representing data as knowledge graphs. We develop a number of algorithms that augment data via processing graph-based information.
We show how these enriched structures can be utilized for further processing, and discovering of `unseen' relations. In particular, we focus on two very different
In particular, we focus on two very different domains of interest: power systems, and aspect-based sentiment analysis.
For a power system application, we construct a knowledge graph that combines topological data with information about technical details of electrical components, as well as past events that occurred in the system. We utilize the graph for extracting details needed for the analysis of different events. We develop methods that enable us to analyze an impact of events on different parts of the system.
For a case of aspect-based sentiment analysis, we are interested in emotional-based representation of reviews. Firstly, we identify words and simple phrases that describe
different aspects of the reviewed items. Secondly, we create a knowledge graph that combines the reviews with the description words and phrases, as well as with the Hourglass model of emotions. The graph allows us to identify emotions linked with the reviewed aspects. Further, we aggregate these emotions across all words and phrases that describe individual aspects to obtain an `emotional summary' of the reviews.
- Graduation date
- Spring 2021
- Type of Item
- Master of Science
- This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.