nlp-reference Goto Github PK

autoeter

EMNLP 2020 Findings: AutoETER: Automated Entity Type Representation for Knowledge Graph Embedding

complex

Source code for experiments in the papers "Complex Embeddings for Simple Link Prediction" (ICML 2016) and "Knowledge Graph Completion via Complex Tensor Factorization" (JMLR 2017).

entity_type_constraint

inductive-relation-prediction-by-subgraph-reasoning

Inductive relation prediction by subgraph reasoning, ICML'20

kemp

[AAAI 2022] Knowledge Bridging for Empathetic Dialogue Generation

knowledgegraphembedding

DistMult Complex TransE

morse

Source code for the SIGIR 2022 paper "Meta-Knowledge Transfer for Inductive Knowledge Graph Embedding"

pathcon

Combining relational context and relational paths for knowledge graph completion

quate

Pytorch implementation of Quaternion Knowledge Graph Embeddings.

the-code-of-paper-incorporating-attributes-semantics-into-knowledge-graph-embeddings-

More and more works have focused on incorporating different kinds of literals into Knowledge Graph to promote the performance of knowledge embedding. These literals contain numeric literals, text literals, image literals and so on. These additional descriptions are connected to the entities through certain attributes. To incorporate numeric literals, some methods combine the embeddings of literals part with the traditional part - embeddings of entities. However, in the construction of literals embeddings, these existing methods consider the differences of these attributes: one dimension represents one attribute. But they ignore semantic meanings of attributes themselves. In this paper, we propose two methods to incorporate attributes semantics into knowledge graph embeddings from two perspectives: LiteralE-AN and literalE-AT. They concatenate with the embeddings of numeric literals by different ways. Furthermore, their extension model LiteralE-C is also proposed have a more comprehensive representation of attributes semantics. In an empirical study over two standard datasets FB15k and FB15k-237, we evaluate our models for link prediction. We demonstrate they show effective way to improve LiteralE and achieve the state-of-the-art results. In ablation experiments, we find combined models do better than their singular counterparts in most cases.

transe