Papers with deep-learning-based sequence labeling model
MyFixit: An Annotated Dataset, Annotation Tool, and Baseline Methods for Information Extraction from Repair Manuals (2020.lrec-1)
Copied to clipboard
| Challenge: | Textual instructions are among the most widely used media for learning and teaching . however, to create autonomous systems, it is difficult to extract task knowledge from text . |
| Approach: | They propose methods that can extract information from repair manuals from a semi-structured dataset . they propose a bag-of-n-grams similarity method and deep-learning-based sequence labeling model . |
| Outcome: | The proposed methods can extract the needed tools and disassembled parts from repair manuals. |