Corpus ID: 2687019

Relation Extraction with Matrix Factorization and Universal Schemas

@inproceedings{Riedel2013RelationEW,
  title={Relation Extraction with Matrix Factorization and Universal Schemas},
  author={S. Riedel and Limin Yao and A. McCallum and Benjamin M Marlin},
  booktitle={HLT-NAACL},
  year={2013}
}
  • S. Riedel, Limin Yao, +1 author Benjamin M Marlin
  • Published in HLT-NAACL 2013
  • Computer Science
  • © 2013 Association for Computational Linguistics. Traditional relation extraction predicts relations within some fixed and finite target schema. Machine learning approaches to this task require either manual annotation or, in the case of distant supervision, existing structured sources of the same schema. The need for existing datasets can be avoided by using a universal schema: the union of all involved schemas (surface form predicates as in OpenIE, and relations in the schemas of preexisting databases). This schema has an almost unlimited set of relations (due to surface forms), and supports integration with existing structured data (through the relation types of existing databases). [...] Key Method To populate a database of such schema we present matrix factorization models that learn latent feature vectors for entity tuples and relations. We show that such latent models achieve substantially higher accuracy than a traditional classification approach. More importantly, by operating simultaneously on relations observed in text and in pre-existing structured DBs such as Freebase, we are able to reason about unstructured and structured data in mutually-supporting ways. By doing so our…Expand Abstract
    552 Citations
    Matrix Models with Feature Enrichment for Relation Extraction
    A Hierarchical Model for Universal Schema Relation Extraction
    • 1
    • Highly Influenced
    • PDF
    Towards Combined Matrix and Tensor Factorization for Universal Schema Relation Extraction
    • 18
    • PDF
    Feature-enriched matrix factorization for relation extraction
    • 7
    • Highly Influenced
    • PDF
    Open Relation Extraction by Matrix Factorization and Universal Schemas
    • Highly Influenced
    • PDF
    Injecting Logical Background Knowledge into Embeddings for Relation Extraction
    • 180
    • PDF
    Universal schema for entity type prediction
    • 44
    • PDF
    Multilingual Relation Extraction using Compositional Universal Schema
    • 74
    • PDF
    Row-less Universal Schema
    • 20
    • PDF
    Weakly-supervised Relation Extraction by Pattern-enhanced Embedding Learning
    • 24
    • PDF

    References

    SHOWING 1-10 OF 35 REFERENCES
    Distant supervision for relation extraction without labeled data
    • 2,061
    • Highly Influential
    • PDF
    Probabilistic Databases of Universal Schema
    • 19
    • PDF
    Multi-instance Multi-label Learning for Relation Extraction
    • 590
    • Highly Influential
    • PDF
    Modeling Relations and Their Mentions without Labeled Text
    • 816
    • PDF
    Structured Relation Discovery using Generative Models
    • 129
    • PDF
    Factorizing YAGO: scalable machine learning for linked data
    • 339
    • PDF
    Open Information Extraction from the Web
    • 2,117
    • PDF
    Unsupervised Methods for Determining Object and Relation Synonyms on the Web
    • 113
    • PDF