Knowledge Detection and Discovery using Semantic Graph Embeddings on Large Knowledge Graphs generated on Text Mining Results

Jens Dörpinghaus,Marc Jacobs

doi:10.15439/2020f36

Abstract

Knowledge graphs play a central role in big data integration, especially for connecting data from different domains. Bringing unstructured texts, e.g. from scientific literature, into a structured, comparable format is one of the key assets. Here, we use knowledge graphs in the biomedical domain working together with text mining based document data for knowledge extraction and retrieval from text and natural language structures. For example cause and effect models, can potentially facilitate clinical decision making or help to drive research towards precision medicine. However, the power of knowledge graphs critically depends on context information. Here we provide a novel semantic approach towards a context enriched biomedical knowledge graph utilizing data integration with linked data applied to language technologies and text mining. This graph concept can be used for graph embedding applied in different approaches, e.g with focus on topic detection, document clustering and knowledge discovery. We discuss algorithmic approaches to tackle these challenges and show results for several applications like search query finding and knowledge discovery. The presented remarkable approaches lead to valuable results on large knowledge graphs.

Highlights

I N THIS paper we will present a novel approach towards knowledge detection and discovery using semantic graph embeddings on large knowledge graphs
For example biological relations might be associated with an ontology, they can be annotated to a document with named entity recognition (NER, annotation layer) and they might belong to a domain specific language layer
In [3] we proved this concept for one single layer containing keywords

Summary

Introduction

I N THIS paper we will present a novel approach towards knowledge detection and discovery using semantic graph embeddings on large knowledge graphs. The idea of semantic graph embeddings was initially introduced in [1], the theoretical background in [2] and the algorithms which are used as a basis for our approach were introduced in [3]. Combining these results, we will present a novel heuristic approach and present experimental results on a large scale knowledge graph from the biomedical field, see [4]. We will present a novel heuristic approach and present experimental results on a large scale knowledge graph from the biomedical field, see [4] This graph is build upon text mining results on biomedical literature databases.

Objectives

Methods

Results

Conclusion