PaperlayerView exact revision on arXiv

Source-linked AI summary

Semantic Relation Classification via Convolutional Neural Networks with Simple Negative Sampling

Kun Xu, Yansong Feng, Songfang Huang, Dongyan Zhao

arXiv:1506.07650v1cs.CLcs.LG

TL;DR

Relation extraction models can be distracted by irrelevant clauses when nominals are distant, motivating representations focused on their shortest dependency path. The paper uses a CNN over these paths and negative sampling for subject–object directionality, achieving better reported performance than prior systems on SemEval-2010 Task 8.

  • Problem

    Neural relation-extraction models often suffer from irrelevant information when subjects and objects are separated by longer distances, while subject–object directionality also requires explicit handling.

    Source: §1 · p. 1

  • Method

    A convolutional neural network learns relation representations from shortest dependency paths, while reversed object-to-subject paths provide negative samples for learning subject–object assignments.

    Source: §1 · p. 1, §3 · p. 2

  • Results

    The model outperforms state-of-the-art systems on SemEval-2010 Task 8, while negative sampling improves F1 by 2.1% without lexical features and 1.9% with them.

    Source: §6 · p. 4, §5 · p. 4

  • Takeaways & Limitations

    Shortest dependency paths capture syntactic features for relation extraction, and simple negative samples help the model make correct subject–object assignments.

    Source: §1 · p. 1, §6 · p. 4

    The paper omits detailed objective-function formulas for space and assumes the shortest path describes the relationship between two nominals in a sentence.

    Source: §2 · p. 2, §4 · p. 3

Abstract

from arXiv · show

Syntactic features play an essential role in identifying relationship in a sentence. Previous neural network models often suffer from irrelevant information introduced when subjects and objects are in a long distance. In this paper, we propose to learn more robust relation representations from the shortest dependency path through a convolution neural network. We further propose a straightforward negative sampling strategy to improve the assignment of subjects and objects. Experimental results show that our method outperforms the state-of-the-art methods on the SemEval-2010 Task 8 dataset.

1 Introduction

The paper targets relation extraction with syntactic representations that avoid irrelevant clauses and explicitly model subject–object directionality. It uses shortest dependency paths and negative sampling to improve relation classification.

  • Relation extraction identifies the relationship between two nominals in a sentence, commonly using syntactic and other linguistic features.Source: §1 · p. 1
  • Neural models can incorporate irrelevant subsequences or clauses when subjects and objects are far apart, hurting extraction performance.Source: §1 · p. 1
  • The proposed CNN learns relation representations from the simple dependency path between subjects and objects, excluding unrelated chunks or clauses.Source: §1 · p. 1, Figure 1 · p. 2
  • Negative sampling models relation directionality by using the shortest dependency path from the object to the subject as a negative sample.Source: §1 · p. 1
  • The model outperforms state-of-the-art methods on the SemEval-2010 Task 8 dataset.Source: §1 · p. 1

2 The Shortest Path Hypothesis

The shortest dependency path is assumed to describe the relationship between two nominals, while predicate order on that path indicates subject–object assignments.

  • For two nominals in one sentence, the shortest dependency path is assumed to describe their relationship.Source: §2 · p. 2
  • If the nominals share a predicate, the path passes through it; across predicate–argument structures, it passes through predicates sharing consecutive arguments.Source: §2 · p. 2
  • The order of predicates on the path indicates the proper assignments of subjects and objects.Source: §2 · p. 2

3 A Convolutional Neural Network Model

The CNN converts shortest dependency paths into fixed-size syntactic representations by embedding path nodes, extracting local features, pooling globally, and classifying relations.

  • The model inputs words, dependency edge directions, and dependency labels from the shortest path between subject and object.Source: §3 · p. 2
  • Each path node is mapped to a vector through an embedding lookup table.Source: §3 · p. 2

Convolution

  • Fixed-size windows concatenate node vectors to form local feature matrices around each dependency-path node.Source: §3 · p. 2
  • A linear transformation extracts local contextual features, and max pooling selects useful features into a global vector independent of path length.Source: §3 · p. 2

Objective Function and Learning

  • The softmax classifier predicts a K-class relation distribution using cross-entropy error and regularization over the training data.Source: §3 · p. 3
  • Model parameters are learned through backpropagation using stochastic gradient descent with AdaGrad.Source: §3 · p. 3

4 Negative Sampling

The model uses negative sampling to learn correct subject–object assignments, treating reversed assignments as informative negative examples. Development experiments show that assignment information materially improves performance.

Pilot Experiments

  • The third pilot experiment serves as an upper bound because subject–object assignments are available during both training and testing.Source: §4 · p. 3

Negative Sampling

  • A dependency path may correspond to multiple relations between two nominals.Source: §4 · p. 3
  • The paper omits detailed formulas because of space limitations.Source: §4 · p. 3
  • Adding assignment information during training significantly improves the model, demonstrating that subject–object assignment is crucial to performance.Source: §4 · p. 3
  • The proposed strategy treats the opposite subject–object assignment as a negative example for the same relationship.The reversed dependency path differs from the correct path, providing information for distinguishing subject and object roles.Source: §4 · p. 3

5 Experimental Evaluation

The model is evaluated on SemEval-2010 Task 8 against established systems and negative-sampling alternatives. It outperforms prior approaches, benefits from dependency labels, and gains further from reversed-path negative sampling and lexical features.

Dataset and Setup

  • 10,717 annotated examples comprise SemEval-2010 Task 8, with 8,000 training instances, 2,717 test instances, and 2,182 validation samples.Source: §5 · p. 3
  • The model predicts both opposite subject–object dependency paths and selects other only when both predictions are other.Source: §5 · p. 3

Main Results

Negative Sampling

Main Results

  • The vanilla depLCNN+NS outperforms prior best systems without extra lexical features, while similar lexical features improve it by 1.6%.Source: §5 · p. 4
  • depCNN and depLCNN outperform MVRNN and CNN by at least 2.2%, and dependency labels make depLCNN stronger than depCNN.The results indicate that dependency-label information provides more discriminative information for relation extraction.Source: §5 · p. 4

Negative Sampling

  • Negative sampling improves F1 by 2.1% without lexical features and by 1.9% with extra lexical features.Source: §5 · p. 4
  • Reversed dependency paths achieve 85.4% F1, improving random negative samples by 1.9% with a similar number of examples.The reversed paths provide clues for learning correct subject–object assignments.Source: §5 · p. 4

6 Conclusion

The paper combines shortest-dependency-path convolutional representations with simple negative sampling for relation extraction. Experiments show significant gains over state-of-the-art systems and effective capture of syntactic features.

Conclusion

  • The model learns relation representations from shortest dependency paths using a convolutional neural network.Source: §6 · p. 4
  • Simple negative sampling helps the model make correct subject–object assignments within relationships.Source: §6 · p. 4
  • Experimental results show significant improvements over state-of-the-art systems and effective capture of syntactic features.Source: §6 · p. 4

Related papers

Loading 1506.07650v1…