Knowledge graph embedding for data mining vs. knowledge graph embedding for link prediction – two sides of the same coin?

Abstract

Knowledge Graph Embeddings, i.e., projections of entities and relations to lower dimensional spaces, have been proposed for two purposes: (1) providing an encoding for data mining tasks, and (2) predicting links in a knowledge graph. Both lines of research have been pursued rather in isolation from each other so far, each with their own benchmarks and evaluation methodologies. In this paper, we argue that both tasks are actually related, and we show that the first family of approaches can also be used for the second task and vice versa. In two series of experiments, we provide a comparison of both families of approaches on both tasks, which, to the best of our knowledge, has not been done so far. Furthermore, we discuss the differences in the similarity functions evoked by the different embedding approaches.

Walk this Way! Entity Walks and Property Walks for RDF2vec

“…3 All embedding models are publicly available to download via KGvec2go [5]. 4 For evaluation, we use the framework proposed in [3], which consists of different tasks (classification, regression, clustering, analogy reasoning, entity relatedness, document similarity). We use a recent DBpedia release 5 .…”

“…It extracts sequences of entities from knowledge graphs, which are then fed into a word2vec encoder [2]. Such embeddings have been shown to be useful in downstream tasks which require numeric representations of entities and rely on a distance metric between entities that captures entity similarity and/or relatedness [4].…”

Refining Diagnosis Paths for Medical Diagnosis based on an Augmented Knowledge Graph

“…In the future, we want to explore the utility of other embedding models beyond RDF2vec [21]. From a technical perspective, other solutions are also possible.…”

“…The prediction itself is then modeled as a binary classification task, using the concatenated vectors of the condition and the rule as input, as described in [21].…”

“…In this paper, we show how patient data can help training a machine learning model to complete the diagnosis paths in the graph. To that end, we explore the use of knowledge graph embeddings [21,30] to predict additional edges in the graph.…”