Efficient and Expressive Knowledge Base Completion Using Subgraph Feature Extraction

  title={Efficient and Expressive Knowledge Base Completion Using Subgraph Feature Extraction},
  author={Matt Gardner and Tom Michael Mitchell},
  booktitle={Conference on Empirical Methods in Natural Language Processing},
We explore some of the practicalities of using random walk inference methods, such as the Path Ranking Algorithm (PRA), for the task of knowledge base completion. [] Key Method In addition to being conceptually simpler than PRA, SFE is much more efficient, reducing computation by an order of magnitude, and more expressive, allowing for much richer features than paths between two nodes in a graph. We show experimentally that this technique gives substantially better performance than PRA and its variants…

Figures and Tables from this paper

Title Knowledge base completion using distinct subgraph paths

This paper addresses the limitations of existing works by introducing a new graph-based feature model – Distinct Subgraph Paths (DSP), which uses a richer set of graph features and therefore can predict new relevant facts that neither SFE, nor PRA or its variants can discover by principle.

Knowledge base completion using distinct subgraph paths

This paper addresses the limitations of existing works by introducing a new graph-based feature model - Distinct Subgraph Paths (DSP), which uses a richer set of graph features and therefore can predict new relevant facts that neither SFE, nor PRA or its variants can discover by principle.

A Knowledge Base Completion Model Based on Path Feature Learning

The path feature learning model (PFLM) is proposed, which aims to learn path features from the existing knowledge base and extra parsed corpus and uses these path features to predict new relations to improve the scalability of knowledge inference.

Context-aware Path Ranking for Knowledge Base Completion

Experimental results show that the path features discovered by the Context-aware Path Ranking algorithm not only improve predictive performance but also are more interpretable than existing baselines.

Learning to Efficiently Propagate for Reasoning on Knowledge Graphs

This paper proposes A*Net, an efficient model for path-based reasoning on knowledge graphs, Inspired by the classical A* algorithm for shortest path problems, which prioritizes important nodes and edges at each propagation step, to reduce the time and memory footprint.

Fine-Grained Evaluation of Rule- and Embedding-Based Systems for Knowledge Graph Completion

This work presents a fine-grained evaluation that gives insight into characteristics of the most popular datasets and points out the different strengths and shortcomings of the examined approaches, and combines both families of approaches via ensemble learning.

Path Ranking with Attention to Type Hierarchies

Attentive Path Ranking is introduced, a novel path pattern representation that leverages type hierarchies of entities to both avoid ambiguity and maintain generalization and experiments demonstrate that the proposed model outperforms existing methods on the fact prediction task.

A*Net: A Scalable Path-based Reasoning Approach for Knowledge Graphs

To the best knowledge, A*Net is the first path-based method for knowledge graph reasoning at such a scale, and inspired by the A* algorithm for shortest path problems, which learns a priority function to select important nodes and edges at each iteration.

CAFE: Fact Checking in Knowledge Graphs using Neighborhood-Aware Features

This paper presents an approach to completing KGs based on evaluating candidate triples using a novel set of features, which exploits the highly relational nature of KGs by analyzing the entities and relations surrounding any given pair of entities.

Incorporating Vector Space Similarity in Random Walk Inference over Knowledge Bases

A new technique for combining KB relations and surface text into a single graph representation that is much more compact than graphs used in prior work is presented, and how to incorporate vector space similarity into random walk inference over KBs is described.

Relational retrieval using a combination of path-constrained random walks

A novel learnable proximity measure is described which instead uses one weight per edge label sequence: proximity is defined by a weighted combination of simple “path experts”, each corresponding to following a particular sequence of labeled edges.

Random Walk Inference and Learning in A Large Scale Knowledge Base

It is shown that a soft inference procedure based on a combination of constrained, weighted, random walks through the knowledge base graph can be used to reliably infer new beliefs for theknowledge base.

Programming with personalized pagerank: a locally groundable first-order probabilistic logic

A first-order probabilistic language which is well-suited to approximate "local" grounding: in particular, every query can be approximately grounded with a small graph.

Improving Learning and Inference in a Large Knowledge-Base using Latent Syntactic Cues

For the first time, it is demonstrated that addition of edges labeled with latent features mined from a large dependency parsed corpus of 500 million Web documents can significantly outperform previous PRAbased approaches on the KB inference task.

Typed Tensor Decomposition of Knowledge Bases for Relation Extraction

A tensor decomposition approach for knowledge base embedding that is highly scalable, and is especially suitable for relation extraction by leveraging relational domain knowledge about entity type information, which is significantly faster than previous approaches and better able to discover new relations missing from the database.

Efficient Random Walk Inference with Knowledge Bases

This thesis describes a new relational learning approach based on path-constrained random walks, and demonstrates, with extensive experiments on IR and NLP tasks, how relational learning can be applied at a scale not possible before.

Knowledge vault: a web-scale approach to probabilistic knowledge fusion

The Knowledge Vault is a Web-scale probabilistic knowledge base that combines extractions from Web content (obtained via analysis of text, tabular data, page structure, and human annotations) with prior knowledge derived from existing knowledge repositories that computes calibrated probabilities of fact correctness.

Knowledge base completion via search-based question answering

A way to leverage existing Web-search-based question-answering technology to fill in the gaps in knowledge bases in a targeted way by learning the best set of queries to ask, such that the answer snippets returned by the search engine are most likely to contain the correct value for that attribute.

Knowledge-Based Weak Supervision for Information Extraction of Overlapping Relations

A novel approach for multi-instance learning with overlapping relations that combines a sentence-level extraction model with a simple, corpus-level component for aggregating the individual facts is presented.