Understanding the World: Bridging Multilingual Knowledge and Translation
Room 2.44, 06/12/2018 (11:30-12:15)
Artificial intelligence (AI) or deep learning (DP) technology has driven our world to a knowledge-based digital world. Machine understanding is the core task in the AI-driven natural language processing (NLP) field, where the semantic knowledge learning, representation and application are the most important components, but the most difficult issues, especially for domain-specific applications, such as biomedical, eHealth etc. Combining the general and domain-specific multilingual knowledge and machine translation technology can tackle these issues to some extent.
The goal of this networking session is to bridge researchers and industry partners from knowledge graph/bases, linked data, machine translation and natural language processing etc. to discuss a better solution to more effectively and efficiently discover, mine, represent, apply and explain multilingual semantic knowledge for better understanding the world and assisting humans. This idea is motivated by the current situation of deep neural networks methodology, i.e. it is difficult to (1) explain what the networks have learned and how the learned semantic knowledge used; (2) seamlessly integrate structured knowledge or prior knowledge from knowledge graph/bases into the deep neural networks.
The outcome can be applied to tasks such as question answering, chatbot, multilingual information retrieval etc. For commercialisation, it can be applied to domains like smart customer service, eHealth, fraud detection etc.
Organised by: Jinhua DU (ADAPT Centre, School of Computing, Dublin City University, Ireland, Ireland)
- Artificial Intelligence
- Big Data
- Open Data
- Smart Cities
Comments are closed