An Algorithm that Learns What‘s in a Name
Machine Learning - Special issue on natural language learning
Marker-Passing inference in the scone knowledge-base system
KSEM'06 Proceedings of the First international conference on Knowledge Science, Engineering and Management
| Hi-index | 0.00 |
We present our progress on a fully automated system that extracts procedures from text. The system extracts a structured, plan-like graph that represents the procedure described in the text it was given. We hope such a system can be used to populate knowledge bases with real-world procedures that will fuel procedure representation and reasoning research geared towards everyday applications. The results obtained on the training data indicate that the task, and the method presented are feasible and should be investigated further.