Large Knowledge Graphs Research Articles

Learning structured embeddings of knowledge graphs with generative adversarial framework

Many large knowledge graphs are now available and ready to provide semantically structured information that is regarded as an important resource for question answering and decision support tasks. However, they are built on rigid symbolic frameworks which makes them hard to be used in other intelligent systems. Knowledge graph embedding approaches are gaining increasing attention, which embeds symbolic entities and relations into continuous vector spaces. Such graph embeddings are often learned by training a model to distinguish true triples from negative ones. Unfortunately, the negative triples created by replacing their heads or tails with randomly selected entities are easily identified by the model, which makes them insufficient to train useful models. To this end, we propose a method under a generative adversarial architecture to learn graph embeddings, in which a generative network is trained to provide continually improved “plausible” triples whereas a discriminative network learns to distinguish truth triples from the others by competing with the generator in a two-player minimax game. When arriving at a convergence, the generative network recovers the training data and can be used for knowledge graph completion, while the discriminative network is trained to be a good triple classifier. Extensive experiments demonstrate our method can improve multiple graph embedding models with a significant margin on both link prediction and triple classification tasks.

Ontology Summit 2021 Communiqué: Ontology generation and harmonization

Advances in machine learning and the development of very large knowledge graphs have accompanied a proliferation of ontologies of many types and for many purposes. These ontologies are commonly developed independently, and as a result, it can be difficult to communicate about and between them. To address this difficulty of communication, ontologies and the communities they serve must agree on how their respective terminologies and formalizations relate to each other. The process of coming into accord and agreement is called “harmonization.” The Ontology Summit 2021 examined the overall landscape of ontologies, the many kinds of ontology generation and harmonization, as well as the sustainability of ontologies. The Communiqué synthesizes and summarizes the findings of the summit as well as earlier summits on related issues. One of the major impediments to harmonization is the relatively poor quality of natural language definitions in many ontologies. The summit surveyed the state of the art in natural language definition development, based on lexicographic principles, as well as examples of ongoing projects that are explicitly dealing with harmonization and sustainability.

KGVQL: A knowledge graph visual query language with bidirectional transformations

With the rapid development of artificial intelligence, knowledge graphs have been widely recognized as a critical component in many AI techniques and systems. A complex knowledge graph may contain hundreds of millions of nodes and edges, thus is challenging for end-users to understand and query. In this paper, we present a knowledge graph interactive visual query language, KGVQL, to improve the efficiency of end-users’ understanding and querying of knowledge graphs. Furthermore, KGVQL realizes the novel capability of flexible bidirectional transformations between query graphs and query results, therefore significantly assisting end-users in constructing queries over large and unfamiliar knowledge graphs in an incremental way. We present the visual syntax of KGVQL, discuss our design rationale behind this interactive visual query language, and illustrate a number of case studies. We empirically evaluate the effectiveness of a visual query system based on KGVQL against a number of textual and visual query environments over a large knowledge graph, DBpedia. Our evaluation demonstrates the superiority of KGVQL in effectiveness and accuracy.

The Microsoft Academic Knowledge Graph enhanced: Author name disambiguation, publication classification, and embeddings

Abstract Although several large knowledge graphs have been proposed in the scholarly field, such graphs are limited with respect to several data quality dimensions such as accuracy and coverage. In this article, we present methods for enhancing the Microsoft Academic Knowledge Graph (MAKG), a recently published large-scale knowledge graph containing metadata about scientific publications and associated authors, venues, and affiliations. Based on a qualitative analysis of the MAKG, we address three aspects. First, we adopt and evaluate unsupervised approaches for large-scale author name disambiguation. Second, we develop and evaluate methods for tagging publications by their discipline and by keywords, facilitating enhanced search and recommendation of publications and associated entities. Third, we compute and evaluate embeddings for all 239 million publications, 243 million authors, 49,000 journals, and 16,000 conference entities in the MAKG based on several state-of-the-art embedding techniques. Finally, we provide statistics for the updated MAKG. Our final MAKG is publicly available at https://makg.org and can be used for the search or recommendation of scholarly entities, as well as enhanced scientific impact quantification.

Efficient semantic summary graphs for querying large knowledge graphs

Knowledge Graphs (KGs) integrate heterogeneous data, but one challenge is the development of efficient tools for allowing end users to extract useful insights from these sources of knowledge. In such a context, reducing the size of a Resource Description Framework (RDF) graph while preserving all information can speed up query engines by limiting data shuffle, especially in a distributed setting. This paper presents two algorithms for RDF graph summarization: Grouping Based Summarization (GBS) and Query Based Summarization (QBS). The latter is an optimized and lossless approach for the former method. We empirically study the effectiveness of the proposed lossless RDF graph summarization to retrieve complete data, by rewriting an RDF Query Language called SPARQL query with fewer triple patterns using a semantic similarity. We conduct our experimental study in instances of four datasets with different sizes. Compared with the state-of-the-art query engine Sparklify executed over the original RDF graphs as a baseline, QBS query execution time is reduced by up to 80% and the summarized RDF graph is decreased by up to 99%.

LinkExplorer: predicting, explaining and exploring links in large biomedical knowledge graphs.

Machine learning algorithms for link prediction can be valuable tools for hypothesis generation. However, many current algorithms are black boxes or lack good user interfaces that could facilitate insight into why predictions are made. We present LinkExplorer, a software suite for predicting, explaining and exploring links in large biomedical knowledge graphs. LinkExplorer integrates our novel, rule-based link prediction engine SAFRAN, which was recently shown to outcompete other explainable algorithms and established black-box algorithms. Here, we demonstrate highly competitive evaluation results of our algorithm on multiple large biomedical knowledge graphs, and release a web interface that allows for interactive and intuitive exploration of predicted links and their explanations. A publicly hosted instance, source code and further documentation can be found at https://github.com/OpenBioLink/Explorer. Supplementary data are available at Bioinformatics online.

Joint extraction of entities and relations by entity role recognition

Joint extracting entities and relations from unstructured text is a fundamental task in information extraction and a key step in constructing large knowledge graphs, entities and relations are constructed as relational triples of the form (subject, relation, object) or (s, r, o). Although triple extraction has been extremely successful, there are still continuing challenges due to factors such as entity overlap. Recent work has shown us the excellent performance of joint extraction models, however these methods still suffer from some problems, such as the redundancy prediction problem. Traditional methods for solving the overlap problem require triple extraction under the full class of relations defined in the dataset, however the number of relations in a sentence is much smaller than the full relational class, which leads to a large number of redundant predictions. To solve this problem, this paper decomposes the task into two steps: entity and potential relation extraction and entity-semantic role determination of triples. Specifically, we design several modules to extract the entities and relations in the sentence separately, and we use these entities and relations to construct possible candidate triples and predict the semantic roles (subject or object) of the entities under the relational constraints to obtain the correct triples. In general we propose a model for identifying the semantic roles of entities in triples under relation constraints, which can effectively solve the problem of redundant prediction, We also evaluated our model on two widely used public datasets, and our model achieved advanced performance with F1 scores of 90.8 and 92.4 on NYT and WebNLG, respectively.

Towards the next generation of the LinkedGeoData project using virtual knowledge graphs

With the advancement of Semantic Technologies, large geospatial data sources have been increasingly published as Linked data on the Web. The LinkedGeoData project is one of the most prominent such projects to create a large knowledge graph from OpenStreetMap (OSM) with global coverage and interlinking of other data sources. In this paper, we report on the ongoing effort of exposing the relational database in LinkedGeoData as a SPARQL endpoint using Virtual Knowledge Graph (VKG) technology. Specifically, we present two realizations of VKGs, using the two systems Sparqlify and Ontop. In order to improve compliance with the OGC GeoSPARQL standard, we have implemented GeoSPARQL support in Ontop v4. Moreover, we have evaluated the VKG-powered LinkedGeoData in the test areas of Italy and Germany. Our experiments demonstrate that such system supports complex GeoSPARQL queries, which confirms that query answering in the VKG approach is efficient.

LargeEA

Entity alignment (EA) aims to find equivalent entities in different knowledge graphs (KGs). Current EA approaches suffer from scalability issues, limiting their usage in real-world EA scenarios. To tackle this challenge, we propose LargeEA to align entities between large-scale KGs. LargeEA consists of two channels, i.e., structure channel and name channel. For the structure channel, we present METIS-CPS, a memory-saving mini-batch generation strategy, to partition large KGs into smaller mini-batches. LargeEA, designed as a general tool, can adopt any existing EA approach to learn entities' structural features within each mini-batch independently. For the name channel, we first introduce NFF, a name feature fusion method, to capture rich name features of entities without involving any complex training process; we then exploit a name-based data augmentation to generate seed alignment without any human intervention. Such design fits common real-world scenarios much better, as seed alignment is not always available. Finally, LargeEA derives the EA results by fusing the structural features and name features of entities. Since no widely-acknowledged benchmark is available for large-scale EA evaluation, we also develop a large-scale EA benchmark called DBP1M extracted from real-world KGs. Extensive experiments confirm the superiority of LargeEA against state-of-the-art competitors.

ABSTAT-HD: a scalable tool for profiling very large knowledge graphs

Processing large-scale and highly interconnected Knowledge Graphs (KG) is becoming crucial for many applications such as recommender systems, question answering, etc. Profiling approaches have been proposed to summarize large KGs with the aim to produce concise and meaningful representation so that they can be easily managed. However, constructing profiles and calculating several statistics such as cardinality descriptors or inferences are resource expensive. In this paper, we present ABSTAT-HD, a highly distributed profiling tool that supports users in profiling and understanding big and complex knowledge graphs. We demonstrate the impact of the new architecture of ABSTAT-HD by presenting a set of experiments that show its scalability with respect to three dimensions of the data to be processed: size, complexity and workload. The experimentation shows that our profiling framework provides informative and concise profiles, and can process and manage very large KGs.

Synospecies, an application to reflect changes in taxonomic names based on a triple store based on taxonomic data liberated from publication

Taxonomic treatments, sections of publications documenting the features or distribution of a related group of organisms (called a “taxon”, plural “taxa”) in ways adhering to highly formalized conventions, and published in scientific journals, shape our understanding of global biodiversity (Catapano 2019). Treatments are the building blocks of the evolving scientific consensus on taxonomic entities. The semantics of these treatments and their relationships are highly structured: taxa are introduced, merged, made obsolete, split, renamed, associated with specimens and so on. Plazi makes this content available in machine-readable form using Resource Description Framework (RDF) . RDF is the standard model for Linked Data and the Semantic Web. RDF can be exchanged in different formats (aka concrete syntaxes) such as RDF/XML or Turtle. The data model describes graph structures and relies on Internationalized Resource Identifiers (IRIs) , ontologies such as Darwin Core basic vocabulary are used to assign meaning to the identifiers. For Synospecies, we unite all treatments into one large knowledge graph, modelling taxonomic knowledge and its evolution with complete references to quotable treatments. However, this knowledge graph expresses much more than any individual treatment could convey because every referenced entity is linked to every other relevant treatment. On synospecies.plazi.org, we provide a user-friendly interface to find the names and treatments related to a taxon. An advanced mode allows execution of queries using the SPARQL query language.

Computing how-provenance for SPARQL queries via query rewriting

Over the past few years, we have witnessed the emergence of large knowledge graphs built by extracting and combining information from multiple sources. This has propelled many advances in query processing over knowledge graphs, however the aspect of providing provenance explanations for query results has so far been mostly neglected. We therefore propose a novel method, SPARQLprov, based on query rewriting, to compute how-provenance polynomials for SPARQL queries over knowledge graphs. Contrary to existing works, SPARQLprov is system-agnostic and can be applied to standard and already deployed SPARQL engines without the need of customized extensions. We rely on spm-semirings to compute polynomial annotations that respect the property of commutation with homomorphisms on monotonic and non-monotonic SPARQL queries without aggregate functions. Our evaluation on real and synthetic data shows that SPARQLprov over standard engines incurs an acceptable runtime overhead w.r.t. the original query, competing with state-of-the-art solutions for how-provenance computation.

Current Status of Baricitinib as a Repurposed Therapy for COVID-19.

The emergence of the COVID-19 pandemic has mandated the instant (re)search for potential drug candidates. In response to the unprecedented situation, it was recognized early that repurposing of available drugs in the market could timely save lives, by skipping the lengthy phases of preclinical and initial safety studies. BenevolentAI’s large knowledge graph repository of structured medical information suggested baricitinib, a Janus-associated kinase inhibitor, as a potential repurposed medicine with a dual mechanism; hindering SARS-CoV2 entry and combatting the cytokine storm; the leading cause of mortality in COVID-19. However, the recently-published Adaptive COVID-19 Treatment Trial-2 (ACTT-2) positioned baricitinib only in combination with remdesivir for treatment of a specific category of COVID-19 patients, whereas the drug is not recommended to be used alone except in clinical trials. The increased pace of data output in all life sciences fields has changed our understanding of data processing and manipulation. For the purpose of drug design, development, or repurposing, the integration of different disciplines of life sciences is highly recommended to achieve the ultimate benefit of using new technologies to mine BIG data, however, the final say remains to be concluded after the drug is used in clinical practice. This review demonstrates different bioinformatics, chemical, pharmacological, and clinical aspects of baricitinib to highlight the repurposing journey of the drug and evaluates its placement in the current guidelines for COVID-19 treatment.

Learning from History: Modeling Temporal Knowledge Graphs with Sequential Copy-Generation Networks

Large knowledge graphs often grow to store temporal facts that model the dynamic relations or interactions of entities along the timeline. Since such temporal knowledge graphs often suffer from incompleteness, it is important to develop time-aware representation learning models that help to infer the missing temporal facts. While the temporal facts are typically evolving, it is observed that many facts often show a repeated pattern along the timeline, such as economic crises and diplomatic activities. This observation indicates that a model could potentially learn much from the known facts appeared in history. To this end, we propose a new representation learning model for temporal knowledge graphs, namely CyGNet, based on a novel time-aware copy-generation mechanism. CyGNet is not only able to predict future facts from the whole entity vocabulary, but also capable of identifying facts with repetition and accordingly predicting such future facts with reference to the known facts in the past. We evaluate the proposed method on the knowledge graph completion task using five benchmark datasets. Extensive experiments demonstrate the effectiveness of CyGNet for predicting future facts with repetition as well as de novo fact prediction.

SumGNN: multi-typed drug interaction prediction via efficient knowledge graph summarization.

Thanks to the increasing availability of drug-drug interactions (DDI) datasets and large biomedical knowledge graphs (KGs), accurate detection of adverse DDI using machine learning models becomes possible. However, it remains largely an open problem how to effectively utilize large and noisy biomedical KG for DDI detection. Due to its sheer size and amount of noise in KGs, it is often less beneficial to directly integrate KGs with other smaller but higher quality data (e.g. experimental data). Most of existing approaches ignore KGs altogether. Some tries to directly integrate KGs with other data via graph neural networks with limited success. Furthermore most previous works focus on binary DDI prediction whereas the multi-typed DDI pharmacological effect prediction is more meaningful but harder task. To fill the gaps, we propose a new method SumGNN: knowledge summarization graph neural network, which is enabled by a subgraph extraction module that can efficiently anchor on relevant subgraphs from a KG, a self-attention based subgraph summarization scheme to generate reasoning path within the subgraph, and a multi-channel knowledge and data integration module that utilizes massive external biomedical knowledge for significantly improved multi-typed DDI predictions. SumGNN outperforms the best baseline by up to 5.54%, and performance gain is particularly significant in low data relation types. In addition, SumGNN provides interpretable prediction via the generated reasoning paths for each prediction. The code is available in Supplementary Material. Supplementary data are available at Bioinformatics online.

KEPLER: A Unified Model for Knowledge Embedding and Pre-trained Language Representation

Abstract Pre-trained language representation models (PLMs) cannot well capture factual knowledge from text. In contrast, knowledge embedding (KE) methods can effectively represent the relational facts in knowledge graphs (KGs) with informative entity embeddings, but conventional KE models cannot take full advantage of the abundant textual information. In this paper, we propose a unified model for Knowledge Embedding and Pre-trained LanguagERepresentation (KEPLER), which can not only better integrate factual knowledge into PLMs but also produce effective text-enhanced KE with the strong PLMs. In KEPLER, we encode textual entity descriptions with a PLM as their embeddings, and then jointly optimize the KE and language modeling objectives. Experimental results show that KEPLER achieves state-of-the-art performances on various NLP tasks, and also works remarkably well as an inductive KE model on KG link prediction. Furthermore, for pre-training and evaluating KEPLER, we construct Wikidata5M1 , a large-scale KG dataset with aligned entity descriptions, and benchmark state-of-the-art KE methods on it. It shall serve as a new KE benchmark and facilitate the research on large KG, inductive KE, and KG with text. The source code can be obtained from https://github.com/THU-KEG/KEPLER.

Entity alignment via knowledge embedding and type matching constraints for knowledge graph inference

The prosperous development of cognition and human-level intelligence have made the creation of large-scale domain knowledge graphs (KGs) a hot research field. Entity alignment (EA), which discovers entity pairs that represent the same real object in different KGs, is the key technology for building large scale knowledge graph. Existing EA methods rely mainly on the vector representation of the entity and ignoring the semantic features of an entity in a KG, result in poor alignment quality. To address this limitation, we propose a new EA framework based on knowledge embeddings (KEs) and type matching constraints. By embedding two KGs into a unified vector space, we combine the similarity of the entity vector and the degree of the entity type matching to perform more accurate cross-KG EA. The experimental results show that our proposed method has very good performance on both cross-KG and cross-language data sets. On the DBP-YG dataset, in terms of Hit@1, Hits@10, and MRR, our methods have an average increase of 27% over the baseline method MTransE. In most cases, our method significantly improves the accuracy of EA compared with state-of-the-art methods.

No One is Perfect: Analysing the Performance of Question Answering Components Over the Dbpedia Knowledge Graph

Question answering (QA) over knowledge graphs has gained significant momentum over the past five years due to the increasing availability of large knowledge graphs and the rising importance of question answering for user interaction. DBpedia has been the most prominently used knowledge graph in this setting and most approaches currently use a pipeline of processing steps connecting a sequence of components. In this article, we analyse and micro evaluate the behaviour of 29 available QA components for the DBpedia knowledge graph that were released by the research community since 2010. As a result, we provide a perspective on collective failure cases, suggest characteristics of QA components that prevent them from performing better and provide future challenges and research directions in the field.

Knowledge Enhanced Fact Checking and Verification

As the Internet and social media offer increasing opportunities for organizations and individuals to publicize online contents, it has become essential to develop effective means to identify misinformation like fake news. Recently, fact checking systems have been regarded as a promising tool to automatically deal with large amounts of information. How to effectively take advantage of existing unstructured document knowledge bases and structured knowledge graphs to build robust fact checking systems, however, remains to be a challenge. In this paper, we propose a knowledge enhanced fact checking system, which leverages the Wikidata5M knowledge graph and Wikipedia documents to incorporate external knowledge into the claim to be checked for more robust and accurate fact checking. First, we devise a contextualized knowledge graph selection method to identify the most relevant sub-graph with the checked claim from the large knowledge graph. We then construct a novel claim-evidence-knowledge graph and use a graph attention network to integrate natural language evidence with structured knowledge triplets by allowing them to propagate information among each other. By integrating the claim, retrieved evidence and selected knowledge triplets in a unified claim-evidence-knowledge graph, our method improves the label accuracy of predicted claims by more than 4% on the FEVER dataset over state-of-the-art fact checking models.

Combining neural networks and symbolic inference in a hybrid cognitive architecture

Recently, there has been a big progress in developing artificial deep-learning neural networks and large-scale knowledge graphs. However, the results in these two research fields have serious drawbacks. The solutions offered by neural networks remain unstable and prone to adversarial attacks: while the percentage of correct answers increases, the incorrect answers often contain glaring errors. Large knowledge graphs contain a lot of facts but little knowledge; they are mostly used for information search and retrieval. The ability to reason conclusions on them is limited, and the majority of modern research turns to approximate methods like neural networks and graph embeddings to draw conclusions and use the accumulated knowledge. In this work, I propose a hybrid cognitive architecture inspired by the observable features of human thinking. Pruning obviously wrong solutions seems to be more natural for human symbolic reasoning than making far-fetched strict logical conclusions, while generating new ideas is often intuitive. So a hybrid cognitive architecture can employ generative neural networks as a sort of "intuition" (generating possible solutions) and symbolic inference as a control contour to verify and filter the generated solutions, weeding out dangerous and wrong ideas. This requires creating knowledge graphs containing negative information: the information of what cannot happen or should not be done and why. The problems of creating negative knowledge graphs are discussed. Hybrid cognitive systems using the proposed architecture will be a lot more trustworthy as they will have a system of human-verifiable rules that ensures avoiding the worst errors which can be used in many fields from decision making to natural-language parsing.