Mining Commonsense Facts from the Physical World
This work addresses the under-explored issue of automatically completing commonsense knowledge bases to improve their coverage for applications relying on such knowledge.
The paper tackles the problem of incomplete coverage in commonsense knowledge bases by proposing a new task to mine commonsense facts from physical world descriptions, resulting in a model that significantly outperforms baselines and the creation of two large annotated datasets with approximately 200k instances each.
Textual descriptions of the physical world implicitly mention commonsense facts, while the commonsense knowledge bases explicitly represent such facts as triples. Compared to dramatically increased text data, the coverage of existing knowledge bases is far away from completion. Most of the prior studies on populating knowledge bases mainly focus on Freebase. To automatically complete commonsense knowledge bases to improve their coverage is under-explored. In this paper, we propose a new task of mining commonsense facts from the raw text that describes the physical world. We build an effective new model that fuses information from both sequence text and existing knowledge base resource. Then we create two large annotated datasets each with approximate 200k instances for commonsense knowledge base completion. Empirical results demonstrate that our model significantly outperforms baselines.