Knowledge graph embedding models have gained significant attention in AI research. Recent works have shown that the inclusion of background knowledge, such as logical rules, can improve the performance of embeddings in downstream machine learning tasks. However, so far, most existing models do not allow the inclusion of rules. We address the challenge of including rules and present a new neural based embedding model (LogicENN). We prove that LogicENN can learn every ground truth of encoded rules in a knowledge graph. To the best of our knowledge, this has not been proved so far for the neural based family of embedding models. Moreover, we derive formulae for the inclusion of various rules, including (anti-)symmetric, inverse, irreflexive and transitive, implication, composition, equivalence and negation. Our formulation allows to avoid grounding for implication and equivalence relations. Our experiments show that LogicENN outperforms the state-of-the-art models in link prediction.

Question answering over knowledge graphs (KGQA) has evolved from simple single-fact questions to complex questions that require graph traversal and aggregation. We propose a novel approach for complex KGQA that uses unsupervised message passing, which propagates confidence scores obtained by parsing an input question and matching terms in the knowledge graph to a set of possible answers. First, we identify entity, relationship, and class names mentioned in a natural language question, and map these to their counterparts in the graph. Then, the confidence scores of these mappings propagate through the graph structure to locate the answer entities. Finally, these are aggregated depending on the identified question type. This approach can be efficiently implemented as a series of sparse matrix multiplications mimicking joins over small local subgraphs. Our evaluation results show that the proposed approach outperforms the state-of-the-art on the LC-QuAD benchmark. Moreover, we show that the performance of the approach depends only on the quality of the question interpretation results, i.e., given a correct relevance score distribution, our approach always produces a correct answer ranking. Our error analysis reveals correct answers missing from the benchmark dataset and inconsistencies in the DBpedia knowledge graph. Finally, we provide a comprehensive evaluation of the proposed approach accompanied with an ablation study and an error analysis, which showcase the pitfalls for each of the question answering components in more detail.

Learning embeddings of entities and relations existing in knowledge bases allows the discovery of hidden patterns in data. In this work, we examine the geometrical space's contribution to the task of knowledge base completion. We focus on the family of translational models, whose performance has been lagging, and propose a model, dubbed HyperKG, which exploits the hyperbolic space in order to better reflect the topological properties of knowledge bases. We investigate the type of regularities that our model can capture and we show that it is a prominent candidate for effectively representing a subset of Datalog rules. We empirically show, using a variety of link prediction datasets, that hyperbolic space allows to narrow down significantly the performance gap between translational and bilinear models.

Natural language understanding (NLU) of text is a fundamental challenge in AI, and it has received significant attention throughout the history of NLP research. This primary goal has been studied under different tasks, such as Question Answering (QA) and Textual Entailment (TE). In this thesis, we investigate the NLU problem through the QA task and focus on the aspects that make it a challenge for the current state-of-the-art technology. This thesis is organized into three main parts: In the first part, we explore multiple formalisms to improve existing machine comprehension systems. We propose a formulation for abductive reasoning in natural language and show its effectiveness, especially in domains with limited training data. Additionally, to help reasoning systems cope with irrelevant or redundant information, we create a supervised approach to learn and detect the essential terms in questions. In the second part, we propose two new challenge datasets. In particular, we create two datasets of natural language questions where (i) the first one requires reasoning over multiple sentences; (ii) the second one requires temporal common sense reasoning. We hope that the two proposed datasets will motivate the field to address more complex problems. In the final part, we present the first formal framework for multi-step reasoning algorithms, in the presence of a few important properties of language use, such as incompleteness, ambiguity, etc. We apply this framework to prove fundamental limitations for reasoning algorithms. These theoretical results provide extra intuition into the existing empirical evidence in the field.

Last week in the first installment of our five-part blog series on AI and graph technology, we gave an overview of four ways graphs add context for artificial intelligence: context for decisions with knowledge graphs, context for efficiency with graph accelerated ML, context for accuracy with connected feature extraction, and context for credibility with AI explainability. This week, we examine knowledge graphs, which provide context for decision support (e.g., for call center staff or support engineers) and help ensure that answers are appropriate to the situation (e.g., autonomous vehicles in rainy driving conditions). This will give you insight into how a graph technology platform like Neo4j enhances AI with knowledge graphs. Knowledge Graphs: Context for Decisions One of the AI areas that's moved into production fastest is decision support. Let's say we're trying to solve a real-world problem: making a decision that requires a human to have the right contextual, relevant information and trying to automate or streamline that process in some way.

Visual Question Answering (VQA) is a challenging task for evaluating the ability of comprehensive understanding of the world. Existing benchmarks usually focus on the reasoning abilities either only on the vision or mainly on the knowledge with relatively simple abilities on vision. However, the ability of answering a question that requires alternatively inferring on the image content and the commonsense knowledge is crucial for an advanced VQA system. In this paper, we introduce a VQA dataset that provides more challenging and general questions about Compositional Reasoning on vIsion and Commonsense, which is named as CRIC. To create this dataset, we develop a powerful method to automatically generate compositional questions and rich annotations from both the scene graph of a given image and some external knowledge graph. Moreover, this paper presents a new compositional model that is capable of implementing various types of reasoning functions on the image content and the knowledge graph. Further, we analyze several baselines, state-of-the-art and our model on CRIC dataset. The experimental results show that the proposed task is challenging, where state-of-the-art obtains 52.26% accuracy and our model obtains 58.38%.

Classical object detection frameworks lack of utilizing objects' surrounding information. In this article, we introduce the graph convolutional networks (GCN) into the object detection, and propose a new framework called OD-GCN (object detection with graph convolutional network). It utilizes the category relationship to improve the detection precision. We set up a knowledge graph to reflect the co-exist relationships among objects. GCN plays the role of post-processing to adjust the output of base object detection models. It is a flexible framework that any pre-trained object detection models can be used as the base model. In the experiments, we try several popular base detection models, OD-GCN always improve mAP by 1-5 pp in COCO dataset. In addition, visualized analysis reveals the benchmark improvement is quite logical in human's opinion.

Knowledge graphs are critical to many enterprises today: They provide the structured data and factual knowledge that drive many products and make them more intelligent and "magical." In general, a knowledge graph describes objects of interest and connections between them. For example, a knowledge graph may have nodes for a movie, the actors in this movie, the director, and so on. Each node may have properties such as an actor's name and age. There may be nodes for multiple movies involving a particular actor. The user can then traverse the knowledge graph to collect information on all the movies in which the actor appeared or, if applicable, directed. Many practical implementations impose constraints on the links in knowledge graphs by defining a schema or ontology. For example, a link from a movie to its director must connect an object of type Movie to an object of type Person. In some cases the links themselves might have their own properties: a link connecting an actor and a movie might have the name of the specific role the actor played. Similarly, a link connecting a politician with a specific role in government might have the time period during which the politician held that role.

Informing professionals about the latest research results in their field is a particularly important task in the field of health care, since any development in this field directly improves the health status of the patients. Meanwhile, social media is an infrastructure that allows public instant sharing of information, thus it has recently become popular in medical applications. In this study, we apply Multi Distance Knowledge Graph Embeddings (MDE) to link physicians and surgeons to the latest medical breakthroughs that are shared as the research results on Twitter. Our study shows that using this method physicians can be informed about the new findings in their field given that they have an account dedicated to their profession. Keywords: Knowledge Graph Embeddings · Social Media · Social Good · Health care · Twitter · Machine Learning 1 Introduction Twitter is a projection of the interactions of a society connected to the internet, which is in constant evolution.

Videolectures.net is planning to present the TheyBuyforYou project in video lectures format. Initially you are kindly invited to read about its first results. We are happy to announce the first release of the knowledge graph for public procurement, integrating tender and company data. Public procurement affects organisations across all sectors. With tenders amounting to close to 2 trillion euros annually in the EU, it is critical that this market operates fairly and efficiently, supporting competitiveness and accountability.