Walk this Way! Entity Walks and Property Walks for RDF2vec

Abstract

RDF2vec is a knowledge graph embedding mechanism which first extracts sequences from knowledge graphs by performing random walks, then feeds those into the word embedding algorithm word2vec for computing vector representations for entities. In this poster, we introduce two new flavors of walk extraction coined e-walks and p-walks, which put an emphasis on the structure or the neighborhood of an entity respectively, and thereby allow for creating embeddings which focus on similarity or relatedness. By combining the walk strategies with order-aware and classic RDF2vec, as well as CBOW and skip-gram word2vec embeddings, we conduct a preliminary evaluation with a total of 12 RDF2vec variants.

The RDF2vec family of knowledge graph embedding methods

“…RDF2vec [57] (and all its variants [50,51]) fall into the category of random walk-based deep learning: Multiple walks are performed within a graph, typically for each node, and the set of walks is then interpreted as sentences by the word2vec language embedding algorithm [31,32]. Conceptually, RDF2vec is similar to node2vec [17] and DeepWalk [43], with the difference that the latter approaches were presented in the context of homogeneous graphs, i.e., graphs with merely one edge type.…”

“…Most recently, entity walks and property walks were presented [51]. Those change the walk generation algorithm in terms of what graph elements are included.…”

“…They concern variations in the walk generation (named e-RDF2vec and p-RDF2vec) as well as training word2vec in an order-aware fashion (named RDF2vec oa ). These extensions have been evaluated on their own on task-based datasets before [50,51]. Preliminary evaluations revealed that the flavor that is chosen influences the weight which is put on different (semantic) features -for example, e-RDF2vec spaces are considered to be more focused on relatedness while there is indication that p-RDF2vec spaces cover fine-grained similarity better.…”

“…While some results of this paper have already been published [50][51][52], the following contributions are novel:…”

Knowledge Graph Embeddings using Neural Ito Process: From Multiple Walks to Stochastic Trajectories

“…Node2vec (Grover and Leskovec, 2016;Ristoski and Paulheim, 2016;Portisch and Paulheim, 2022;Huang et al, 2021;Perozzi et al, 2014) are among the models which perform a biased random walk in graph space and compute the low dimensional representation of nodes in such a way to maximize the likelihood of preserving network neighborhoods of nodes. The random walk-based models are not among state-of-the-art KGE models in the link prediction task.…”