Consolidating Commonsense Knowledge

Abstract

Commonsense reasoning is an important aspect of building robust AI systems and is receiving significant attention in the natural language understanding, computer vision, and knowledge graphs communities. At present, a number of valuable commonsense knowledge sources exist, with different foci, strengths, and weaknesses. In this paper, we list representative sources and their properties. Based on this survey, we propose principles and a representation model in order to consolidate them into a Common Sense Knowledge Graph (CSKG). We apply this approach to consolidate seven separate sources into a first integrated CSKG. We present statistics of CSKG, present initial investigations of its utility on four QA datasets, and list learned lessons.

14 pages

References in corpus (12)

• ConceptNet 5.5: An Open Multilingual Graph of General Knowledge
• CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge
• The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence
• PyTorch-BigGraph: A Large-scale Graph Embedding System
• Stereotyping and Bias in the Flickr30K Dataset
• Recent Advances in Natural Language Inference: A Survey of Benchmarks, Resources, and Approaches
• Machine Common Sense Concept Paper
• Abductive Commonsense Reasoning
• Bridging Knowledge Graphs to Generate Scene Graphs
• Yuanfudao at SemEval-2018 Task 11: Three-way Attention and Relational Knowledge for Commonsense Machine Comprehension
• Towards Generalizable Neuro-Symbolic Systems for Commonsense Question Answering
• What's Missing: A Knowledge Gap Guided Approach for Multi-hop Question Answering

Cited by in corpus (5)

• Bridging Knowledge Graphs to Generate Scene Graphs
• Commonsense Knowledge in Wikidata
• Knowledge-driven Data Construction for Zero-shot Evaluation in Commonsense Question Answering
• Lexically-constrained Text Generation through Commonsense Knowledge Extraction and Injection
• Commonsense Knowledge in Word Associations and ConceptNet