Neuro-Symbolic AI-based Intelligent System for Geoscience and Sustainable Industry: Energy and Manufacturing
Artificial Intelligence (AI) is one of the central pillars of Industry 4.0 and the Internet of Things in the global trend towards smart and highly-digitalised plants. An important approach here is to use neuro-symbolic AI, including semantic digital twins and machine learning methods. The former one is to represent experts’ knowledge, physical equipment and processes as unified conceptual semantic models, allowing intelligent and automated reasoning to extract insights, knowledge and decisions over structured and unstructured data. The latter one is to leverage the power of data science tools, such as (deep) neural networks and other advanced data modelling methods for statistic and biology-inspired pattern mining from data of huge volume, variety to generate estimations or predictions that contribute to industrial value-chains. We collaborate with partners that are giant players in the industry, such as Bosch and renowned energy companies like Equinor, and deliver methodologies and knowledge for domains in Geoscience and Sustainable Energy.
Scope of the Thesis
The purpose of this Master thesis is to develop novel techniques and improve our system for Geoscience and Energy Production. Our system has multiple layers and components. We create knowledge models and structured data from real industry data and design sophisticated systems based on semantic technologies, such as ontologies, rules, knowledge graphs, description logic, reasoning etc. Furthermore, we layer numerical pattern mining components upon the semantic counterparts, such as fuzzy logic, deep learning, classic ML methods, etc.
If you take this thesis, you will read relevant papers on semantic technologies and machine learning, develop methods for knowledge modelling and pattern recognition, and conduct studies on knowledge, methods, and prototypes of intelligent systems that can contribute to geoscientific exploration, lithological composite prediction, energy production analysis, or manufacturing monitoring. If the time permits, we strongly encourage students to contribute to scientific publications at prestigious venues.
Master thesis students will enjoy top quality supervision in an enthusiastic international environment of researchers and industrial professionals from SIRIUS and its partners. We could recommend highly-qualified students for further opportunities, such as internships and PhD positions at our industrial partners.
- We expect you to be comfortable with communication in English both in speaking and writing.
- We expect you to be interested in learning some knowledge in geoscience.
- Recommended prior knowledge: machine learning, or semantic technologies, or geoscience.
- Please contact Baifan Zhou for further information.
DigiWell, PeTWIN, SIndAIS4