AI‐based knowledge generation and representation
computing Knowledge Graphs
for AI Conversational
AI Scalable Training
- Goal: Enable a tighter integration of knowledge-driven and statistical approaches to AI by laying the foundations for a knowledge-aware computing architecture
- Design an initial rule-based language for expressive, recursive view definitions over knowledge graphs
- Investigate the conceptual foundations of integrating other AI methods in this rule-based framework based on well-defined interfaces
- Develop methods for interpreting and validating results of other AI formalisms with respect to the original knowledge graph.
- Create a prototype implementation for these concepts, and evaluate it for performance and utility
- AI-based maintenance and generation of the largest free and open knowledge graph (LOD)
- Export of millions of derived knowledge graphs for individual AI use cases
- Hybrid AI approach for integrating your data as background knowledge into a dialogue system
- Investigation of end-to-end learning approaches trained on raw dialogue data.
- Creation of approaches to help lay users to control expert systems (e.g. robots, data science tool chains)
- Verbalization strategies to ensure a natural formulation of answers
This project studies the aquistion of training data from the web for distant supervision.
- The Web as Corpus
- Almost all aspects of society are represented on the web. Many prediction tasks relate to supporting humans.
- The web is harnessed to fuel machine learning models.
- Distant supervision
- For a given task, one does not necessarily need data that ideally matches the problem.
- Rather, one can get by with loosely matching data.
- Key goal is the semi-automatic search for such data.
- Many a task requires the additional labeling of data.
- Scale can only be attained by outsourcing to the crowd.
- Key goal is the facilitation of crowdsourcing by dedicated tools, and the automation of designing worker interfaces.