Extracting Commonsense Properties from Embeddings with Limited Human Guidance
| Autor: | Lawrence Birnbaum, Ji Ping Wang, Douglas C Downey, Yiben Yang |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Commonsense knowledge
business.industry Computer science Active learning (machine learning) media_common.quotation_subject Intelligent decision support system Common sense 02 engineering and technology 010501 environmental sciences Machine learning computer.software_genre 01 natural sciences 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Artificial intelligence business computer 0105 earth and related environmental sciences media_common |
| Zdroj: | ACL (2) |
| DOI: | 10.18653/v1/p18-2102 |
| Popis: | Intelligent systems require common sense, but automatically extracting this knowledge from text can be difficult. We propose and assess methods for extracting one type of commonsense knowledge, object-property comparisons, from pre-trained embeddings. In experiments, we show that our approach exceeds the accuracy of previous work but requires substantially less hand-annotated knowledge. Further, we show that an active learning approach that synthesizes common-sense queries can boost accuracy. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|