Description of the research topic:
Recent advances in precision agriculture (PA) utilize the newest developments in the field of Artificial Intelligence (AI) and Data Science (DS) such that the good performance of Computer Vision (CV) and Machine Learning (ML) methods, just to name a few. These methods work well in certain cases when the task is, for example, object detection, recognition or simple prediction. To achieve more advanced capabilities in more complex tasks, it would be beneficial to incorporate the knowledge of the user into these methods. It is especially important when not enough data is available for training or only a small subset of it is labeled. This, however, poses several challenges: First, how to get and represent the knowledge of a person what is a task related to knowledge engineering. Second, how to utilize this knowledge within a ML model so that a desirable performance can be reached even in case of learning from small data. The third issue, besides how to get, represent and utilize human knowledge in ML models, is the creation of „new knowledge” and generate unusual settings which might be interesting from the user’s perspective.
The goal of the thesis is to investigate and research various approaches to knowledge-based and decision support systems and develop a framework for knowledge fusion into ML methods with possibility to include the user into the learning and reasoning process.
The developed methods and implemented framework will be tested in the application domain of precision agriculture, an area which is gaining importance recently from ecological, economical as well as social point of view.
Keywords: Knowledge-based systems, decision support, machine learning, simulation, knowledge representation, data science, nature-inspired computing, precision farming.
Thesis supervisor: Tomas Horvath
Required language skills: English
Further requirements: Interest in farming related topics. Good data science knowledge, good programming skills. Familiarity with knowledge-based systems, simulation models and nature-inspired algorithms is an advantage.
How to Apply?
If you are interested apply here: [PhD] Doctoral School of Informatics – Eötvös Loránd University (elte.hu)
For more information visite the following website: Doctoral School of Informatics (elte.hu)
Funded: Not Funded
Master Degree: Required
Duration: 4 Years
Full/Part Time: Full Time
Starting Date: 06 September 2021
Deadline to Apply: 31 May 2021
Only local Hubs members can access this page. Join the community today: https://phdhub.eu/register
Fields of Science: