Commonsense mining as knowledge base completion? A study on the impact of novelty

Commonsense knowledge bases such as ConceptNet represent knowledge in the form of relational triples. Inspired by recent work by Li et al., we analyse if knowledge base completion models can be used to mine commonsense knowledge from raw text. We propose novelty of predicted triples with respect to the training set as an important factor in interpreting results. We critically analyse the difficulty of mining novel commonsense knowledge, and show that a simple baseline method that outperforms the previous state of the art on predicting more novel triples.

• Publication year

  2018

• Venue

  arXiv.org

• Publication date

  2018-04-01

• Fields of study

  Computer Science

• Identifiers
  DOI 10.18653/v1/W18-1002arXiv 1804.09259
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Reasoning

  2019cited by this paper
• Verb Physics: Relative Physical Knowledge of Actions and Objects

  2017cited by this paper
• Convolutional 2D Knowledge Graph Embeddings

  2017cited by this paper
• Knowledge Base Completion: Baselines Strike Back

  2017cited by this paper
• Fast Linear Model for Knowledge Graph Embeddings

  2017cited by this paper
• Commonsense Knowledge Base Completion

  2016cited by this paper
• Never-Ending Learning

  2015cited by this paper
• Do Supervised Distributional Methods Really Learn Lexical Inference Relations?

  2015cited by this paper
• Learning Word Representations by Jointly Modeling Syntagmatic and Paradigmatic Relations

  2015cited by this paper
• Philosophers are Mortal: Inferring the Truth of Unseen Facts

  2013cited by this paper
• Reasoning With Neural Tensor Networks for Knowledge Base Completion

  2013cited by this paper
• Representing General Relational Knowledge in ConceptNet 5

  2012influential reference
• Adaptive Subgradient Methods for Online Learning and Stochastic Optimization

  2011cited by this paper
• A Three-Way Model for Collective Learning on Multi-Relational Data

  2011cited by this paper
• Open Information Extraction from the Web

  2007cited by this paper
• The Word-Space Model: using distributional analysis to represent syntagmatic and paradigmatic relations between words in high-dimensional vector spaces

  2006cited by this paper

• COM-transformers: research on automatic knowledge graph construction based on transformer

  2025cites this paper
• Commonsense Knowledge Graph Completion Via Contrastive Pretraining and Node Clustering

  2023cites this paper
• Dense-ATOMIC: Construction of Densely-connected and Multi-hop Commonsense Knowledge Graph upon ATOMIC

  2022cites this paper
• PseudoReasoner: Leveraging Pseudo Labels for Commonsense Knowledge Base Population

  2022cites this paper
• Commonsense Knowledge Base Completion with Relational Graph Attention Network and Pre-trained Language Model

  2022cites this paper
• Are Visual-Linguistic Models Commonsense Knowledge Bases?

  2022cites this paper
• Probing Commonsense Knowledge in Pre-trained Language Models with Sense-level Precision and Expanded Vocabulary

  2022cites this paper
• Dense-ATOMIC: Towards Densely-connected ATOMIC with High Knowledge Coverage and Massive Multi-hop Paths

  2022cites this paper
• Alleviating the Knowledge-Language Inconsistency: A Study for Deep Commonsense Knowledge

  2021cites this paper
• Commonsense Knowledge Mining from Term Definitions

  2021cites this paper
• Do Language Models Perform Generalizable Commonsense Inference?

  2021cites this paper
• Siamese Pre-Trained Transformer Encoder for Knowledge Base Completion

  2021cites this paper
• Distilling Relation Embeddings from Pretrained Language Models

  2021cites this paper
• Transformer-Based Siamese Textual Encoder for Knowledge Base Completion

  2021cites this paper
• Dynamic Neuro-Symbolic Knowledge Graph Construction for Zero-shot Commonsense Question Answering

  2020cites this paper
• NegatER: Unsupervised Discovery of Negatives in Commonsense Knowledge Bases

  2020influential citation
• Generating Negative Commonsense Knowledge

  2020influential citation
• Paragraph-Level Commonsense Transformers with Recurrent Memory

  2020cites this paper
• Commonsense Knowledge Base Completion with Structural and Semantic Context

  2019cites this paper
• Exploiting Structural and Semantic Context for Commonsense Knowledge Base Completion

  2019influential citation
• Commonsense Knowledge Mining from Pretrained Models

  2019influential citation
• Comprehensive structured knowledge base system construction with natural language presentation

  2019cites this paper
• Recent Advances in Natural Language Inference: A Survey of Benchmarks, Resources, and Approaches

  2019cites this paper
• Dynamic Knowledge Graph Construction for Zero-shot Commonsense Question Answering

  2019cites this paper
• Do Language Models Have Common Sense

  2018influential citation
• Machine Common Sense Concept Paper

  2018cites this paper
• Introductory Tutorial: Commonsense Reasoning for Natural Language Processing

  year unknowncites this paper