CSKG: The CommonSense Knowledge Graph
This work addresses the problem of fragmented commonsense knowledge for researchers building AI systems that require robust understanding and reasoning capabilities.
This paper integrates seven key commonsense knowledge sources into a unified CommonSense Knowledge Graph (CSKG) by following five consolidation principles. Analysis shows CSKG is well-connected and its embeddings are useful for downstream reasoning and pre-training language models.
Sources of commonsense knowledge support applications in natural language understanding, computer vision, and knowledge graphs. Given their complementarity, their integration is desired. Yet, their different foci, modeling approaches, and sparse overlap make integration difficult. In this paper, we consolidate commonsense knowledge by following five principles, which we apply to combine seven key sources into a first integrated CommonSense Knowledge Graph (CSKG). We analyze CSKG and its various text and graph embeddings, showing that CSKG is well-connected and that its embeddings provide a useful entry point to the graph. We demonstrate how CSKG can provide evidence for generalizable downstream reasoning and for pre-training of language models. CSKG and all its embeddings are made publicly available to support further research on commonsense knowledge integration and reasoning.