SAKD: Sustainable Agriculture using Knowledge Graphs and Deep Learning
Assigned to: Armin Hohenegger | Level: Bachelor
This thesis was chosen as one of the finalists for the best student thesis award.
The change in climate, depleting soil quality and the increase in the world population has raised concern about food security. Even the United Nations (UN) has included it as one of the Sustainable Development Goals (SDG) 2030. Agricultural sustainability is required to address this alarming situation and, achieving UN SDG. Disruptive technologies can contribute to agricultural sustainability in the modern era. However, agricultural sustainability is not an easy task due to the complexities of the agricultural ecosystem. Although there have been studies on the use of artificial intelligence (AI) (e.g. machine learning and deep learning) and the Internet of Things (IoT) for smart agriculture, there is a lack of research on Smart Sustainable Agriculture (SSA). Complications such as fragmented agricultural processes, interoperability, and a large volume of generated data add to the complexity, posing an ongoing challenge to sustainable agriculture. Semantic technologies, on the other hand, can be used to transform fragmented raw data into knowledge through knowledge graphs and ontologies. Semantic technology further enables interoperability and can aid machine learning by providing contextual information. Besides, machine learning and deep learning techniques can be used to discover hidden patterns and make predictions based on the discovered patterns. Therefore, this project aims to integrate semantic technology and deep learning techniques, utilising the best of both to address the issue of agricultural sustainability.
Keywords: Sustainability, Decision Systems, Knowledge Graphs, Deep Learning, Smart Agriculture, IoT