No file available
To read the file of this research,
you can request a copy directly from the authors.
Siamese Graph Neural Networks for Data Integration
Preprints and early-stage research may not have been peer reviewed yet.
Abstract
Data integration has been studied extensively for decades and approached from different angles. However, this domain still remains largely rule-driven and lacks universal automation. Recent development in machine learning and in particular deep learning has opened the way to more general and more efficient solutions to data integration problems. In this work, we propose a general approach to modeling and integrating entities from structured data, such as relational databases, as well as unstructured sources, such as free text from news articles. Our approach is designed to explicitly model and leverage relations between entities, thereby using all available information and preserving as much context as possible. This is achieved by combining siamese and graph neural networks to propagate information between connected entities and support high scalability. We evaluate our method on the task of integrating data about business entities, and we demonstrate that it outperforms standard rule-based systems, as well as other deep learning approaches that do not use graph-based representations.
... Hence, further research on improved methods and humanin-the-loop approaches is needed. For example, Graph Neural Networks seem promising to improve the accuracy of structural data integration steps [15]. Since health data is inherently different from many other data domains -e.g., due to it being longitudinal and sometimes of low quality -the applicability of methods developed for non-health data remains to be evaluated. ...
Making health data available for secondary use enables innovative data-driven medical research. Since modern machine learning (ML) methods and precision medicine require extensive amounts of data covering most of the standard and edge cases, it is essential to initially acquire large datasets. This can typically only be achieved by integrating different datasets from various sources and sharing data across sites. To obtain a unified dataset from heterogeneous sources, standard representations and Common Data Models (CDM) are needed. The process of mapping data into these standardized representations is usually very tedious and requires many manual configuration and refinement steps. A potential way to reduce these efforts is to use ML methods not only for data analysis, but also for the integration of health data on the syntactic, structural, and semantic level. However, research on ML-based medical data integration is still in its infancy. In this article, we describe the current state of the literature and present selected methods that appear to have a particularly high potential to improve medical data integration. Moreover, we discuss open issues and possible future research directions.
Recent deep learning approaches for representation learning on graphs follow a neighborhood ag-gregation procedure. We analyze some important properties of these models, and propose a strategy to overcome those. In particular, the range of "neighboring" nodes that a node's representation draws from strongly depends on the graph structure, analogous to the spread of a random walk. To adapt to local neighborhood properties and tasks, we explore an architecture-jumping knowledge (JK) networks-that flexibly leverages, for each node, different neighborhood ranges to enable better structure-aware representation. In a number of experiments on social, bioinformatics and citation networks, we demonstrate that our model achieves state-of-the-art performance. Furthermore, combining the JK framework with models like Graph Convolutional Networks, GraphSAGE and Graph Attention Networks consistently improves those models' performance.
Modeling and generating graphs is fundamental for studying networks in biology, engineering, and social sciences. However, modeling complex distributions over graphs and then efficiently sampling from these distributions is challenging due to the non-unique, high-dimensional nature of graphs and the complex, non-local dependencies that exist between edges in a given graph. Here we propose GraphRNN, a deep autoregressive model that addresses the above challenges and approximates any distribution of graphs with minimal assumptions about their structure. GraphRNN learns to generate graphs by training on a representative set of graphs and decomposes the graph generation process into a sequence of node and edge formations, conditioned on the graph structure generated so far. In order to quantitatively evaluate the performance of GraphRNN, we introduce a benchmark suite of datasets, baselines and novel evaluation metrics based on Maximum Mean Discrepancy, which measure distances between sets of graphs. Our experiments show that GraphRNN significantly outperforms all baselines, learning to generate diverse graphs that match the structural characteristics of a target set, while also scaling to graphs 50 times larger than previous deep models.
Identifying the relationship between two text objects is a core research problem underlying many natural language processing tasks. A wide range of deep learning schemes have been proposed for text matching, mainly focusing on sentence matching, question answering or query document matching. We point out that existing approaches do not perform well at matching long documents, which is critical, for example, to AI-based news article understanding and event or story formation. The reason is that these methods either omit or fail to fully utilize complicated semantic structures in long documents. In this paper, we propose a graph approach to text matching, especially targeting long document matching, such as identifying whether two news articles report the same event in the real world, possibly with different narratives. We propose the Concept Interaction Graph to yield a graph representation for a document, with vertices representing different concepts, each being one or a group of coherent keywords in the document, and with edges representing the interactions between different concepts, connected by sentences in the document. Based on the graph representation of document pairs, we further propose a Siamese Encoded Graph Convolutional Network that learns vertex representations through a Siamese neural network and aggregates the vertex features though Graph Convolutional Networks to generate the matching result. Extensive evaluation of the proposed approach based on two labeled news article datasets created at Tencent for its intelligent news products show that the proposed graph approach to long document matching significantly outperforms a wide range of state-of-the-art methods.
We present NeuroSAT, a message passing neural network that learns to solve SAT problems after only being trained as a classifier to predict satisfiability. Although it is not competitive with state-of-the-art SAT solvers, NeuroSAT can solve problems that are substantially larger and more difficult than it ever saw during training by simply running for more iterations. Moreover, NeuroSAT generalizes to novel distributions; after training only on random SAT problems, at test time it can solve SAT problems encoding graph coloring, clique detection, dominating set, and vertex cover problems, all on a range of distributions over small random graphs.
While named entity recognition is a much addressed research topic, recognizing companies in text is of particular difficulty. Company names are extremely heterogeneous in structure, a given company can be referenced in many different ways, their names include person names, locations, acronyms, numbers, and other unusual tokens. Further, instead of using the official company name, quite different colloquial names are frequently used by the general public. We present a machine learning (CRF) system that reliably recognizes organizations in German texts. In particular, we construct and employ various dictionaries, regular expressions, text context, and other techniques to improve the results. In our experiments we achieved a precision of 91.11% and a recall of 78.82%, showing significant improvement over related work. Using our system we were able to extract 263,846 company mentions from a corpus of 141,970 newspaper articles.
Entity resolution is an important application in field of data cleaning. Standard approaches like deterministic methods and probabilistic methods are generally used for this purpose. Many new approaches using single layer perceptron, crowdsourcing etc. are developed to improve the efficiency and also to reduce the time of entity resolution. The approaches used for this purpose also depend on the type of dataset, labeled or unlabeled. This paper presents a new method for labeled data which uses single layered convolutional neural network to perform entity resolution. It also describes how crowdsourcing can be used with the output of the convolutional neural network to further improve the accuracy of the approach while minimizing the cost of crowdsourcing. The paper also discusses the data pre-processing steps used for training the convolutional neural network. Finally it describes the airplane sensor dataset which is used for demonstration of this approach and then shows the experimental results achieved using convolutional neural network.
Over the past several years the amount of published open data has increased significantly. The majority of this is tabular data, that requires powerful and flexible approaches for data cleaning and preparation in order to convert it into Linked Data. This paper introduces Grafterizer – a software framework developed to support data workers and data developers in the process of converting raw tabular data into linked data. Its main components include Grafter, a powerful software library and DSL for data cleaning and RDF-ization, and Grafterizer, a user interface for interactive specification of data transformations along with a back-end for management and execution of data transformations. The proposed demonstration will focus on Grafterizer’s powerful features for data cleaning and RDF-ization in a scenario using data about the risk of failure of transport infrastructure components due to natural hazards.
Entity matching is the problem of determining if two entities in a data set refer to the same real-world object. In the last decade a growing number of large-scale knowledge bases have been created online. Tools for automatically aligning these sources would make it possible to unify them in a structured knowledge and to answer complex queries. Here we present Holistic Entity Matching (HolisticEM), an algorithm based on Personalized Page Rank for aligning instances in large knowledge bases. It consists of two steps. First, a graph of potential matching pairs is constructed; second, local and global information from the relationship graph is propagated via Personalized Page Rank. We demonstrate that HolisticEM performs competitively and can efficiently handle databases with 110M and 203M entities accurately resolving 1.6M of matching entity pairs.
The DBpedia community project extracts structured, multilingual knowledge from Wikipedia and makes it freely available on the Web using Semantic Web and Linked Data technologies. The project extracts knowledge from 111 different language editions of Wikipedia. The largest DBpedia knowledge base which is extracted from the English edition of Wikipedia consists of over 400 million facts that describe 3.7 million things. The DBpedia knowledge bases that are extracted from the other 110 Wikipedia editions together consist of 1.46 billion facts and describe 10 million additional things. The DBpedia project maps Wikipedia infoboxes from 27 different language editions to a single shared ontology consisting of 320 classes and 1,650 properties. The mappings are created via a world-wide crowd-sourcing effort and enable knowledge from the different Wikipedia editions to be combined. The project publishes releases of all DBpedia knowledge bases for download and provides SPARQL query access to 14 out of the 111 language editions via a global network of local DBpedia chapters. In addition to the regular releases, the project maintains a live knowledge base which is updated whenever a page in Wikipedia changes. DBpedia sets 27 million RDF links pointing into over 30 external data sources and thus enables data from these sources to be used together with DBpedia data. Several hundred data sets on the Web publish RDF links pointing to DBpedia themselves and make DBpedia one of the central interlinking hubs in the Linked Open Data (LOD) cloud. In this system report, we give an overview of the DBpedia community project, including its architecture, technical implementation, maintenance, internationalisation, usage statistics and applications.
The recently introduced continuous Skip-gram model is an efficient method for
learning high-quality distributed vector representations that capture a large
number of precise syntactic and semantic word relationships. In this paper we
present several extensions that improve both the quality of the vectors and the
training speed. By subsampling of the frequent words we obtain significant
speedup and also learn more regular word representations. We also describe a
simple alternative to the hierarchical softmax called negative sampling. An
inherent limitation of word representations is their indifference to word order
and their inability to represent idiomatic phrases. For example, the meanings
of "Canada" and "Air" cannot be easily combined to obtain "Air Canada".
Motivated by this example, we present a simple method for finding phrases in
text, and show that learning good vector representations for millions of
phrases is possible.
We present a new technique called "t-SNE" that visualizes high-dimensional data by giving each datapoint a location in a two or three-dimensional map. The technique is a variation of Stochastic Neighbor Embedding (Hinton and Roweis, 2002) that is much easier to optimize, and produces significantly better visualizations by reducing the tendency to crowd points together in the center of the map. t-SNE is better than existing techniques at creating a single map that reveals structure at many different scales. This is particularly important for high-dimensional data that lie on several different, but related, low-dimensional manifolds, such as images of objects from multiple classes seen from multiple viewpoints. For visualizing the structure of very large data sets, we show how t-SNE can use random walks on neighborhood graphs to allow the implicit structure of all of the data to influence the way in which a subset of the data is displayed. We illustrate the performance of t-SNE on a wide variety of data sets and compare it with many other non-parametric visualization techniques, including Sammon mapping, Isomap, and Locally Linear Embedding. The visualiza-tions produced by t-SNE are significantly better than those produced by the other techniques on almost all of the data sets.
A new graphical display is proposed for partitioning techniques. Each cluster is represented by a so-called silhouette, which is based on the comparison of its tightness and separation. This silhouette shows which objects lie well within their cluster, and which ones are merely somewhere in between clusters. The entire clustering is displayed by combining the silhouettes into a single plot, allowing an appreciation of the relative quality of the clusters and an overview of the data configuration. The average silhouette width provides an evaluation of clustering validity, and might be used to select an ‘appropriate’ number of clusters.
Dimensionality reduction involves mapping a set of high dimensional input points onto a low dimensional manifold so that 'similar" points in input space are mapped to nearby points on the manifold. We present a method - called Dimensionality Reduction by Learning an Invariant Mapping (DrLIM) - for learning a globally coherent nonlinear function that maps the data evenly to the output manifold. The learning relies solely on neighborhood relationships and does not require any distancemeasure in the input space. The method can learn mappings that are invariant to certain transformations of the inputs, as is demonstrated with a number of experiments. Comparisons are made to other techniques, in particular LLE.
In several applications the information is naturally represented by graphs. Traditional approaches cope with graphical data structures using a preprocessing phase which transforms the graphs into a set of flat vectors. However, in this way, important topological information may be lost and the achieved results may heavily depend on the preprocessing stage. This paper presents a new neural model, called graph neural network (GNN), capable of directly processing graphs. GNNs extends recursive neural networks and can be applied on most of the practically useful kinds of graphs, including directed, undirected, labelled and cyclic graphs. A learning algorithm for GNNs is proposed and some experiments are discussed which assess the properties of the model.
We present a method for training a similarity metric from data. The method can be used for recognition or verification applications where the number of categories is very large and not known during training, and where the number of training samples for a single category is very small. The idea is to learn a function that maps input patterns into a target space such that the L1 norm in the target space approximates the "semantic" distance in the input space. The method is applied to a face verification task. The learning process minimizes a discriminative loss function that drives the similarity metric to be small for pairs of faces from the same person, and large for pairs from different persons. The mapping from raw to the target space is a convolutional network whose architecture is designed for robustness to geometric distortions. The system is tested on the Purdue/AR face database which has a very high degree of variability in the pose, lighting, expression, position, and artificial occlusions such as dark glasses and obscuring scarves.
Much of human knowledge is encoded in text, available in scientific publications, books, and the web. Given the rapid growth of these resources, we need automated methods to extract such knowledge into machine-processable structures, such as knowledge graphs. An important task in this process is entity normalization, which consists of mapping noisy entity mentions in text to canonical entities in well-known reference sets. However, entity normalization is a challenging problem; there often are many textual forms for a canonical entity that may not be captured in the reference set, and entities mentioned in text may include many syntactic variations, or errors. The problem is particularly acute in scientific domains, such as biology. To address this problem, we have developed a general, scalable solution based on a deep Siamese neural network model to embed the semantic information about the entities, as well as their syntactic variations. We use these embeddings for fast mapping of new entities to large reference sets, and empirically show the effectiveness of our framework in challenging bio-entity normalization datasets.
As data volume and variety have increased, so have the ties between machine learning and data integration become stronger. For machine learning to be effective, one must utilize data from the greatest possible variety of sources; and this is why data integration plays a key role. At the same time machine learning is driving automation in data integration, resulting in overall reduction of integration costs and improved accuracy. This tutorial focuses on three aspects of the synergistic relationship between data integration and machine learning: (1) we survey how state-of-the-art data integration solutions rely on machine learning-based approaches for accurate results and effective human-in-the-loop pipelines, (2) we review how end-to-end machine learning applications rely on data integration to identify accurate, clean, and relevant data for their analytics exercises, and (3) we discuss open research challenges and opportunities that span across data integration and machine learning.
Entity matching (EM) finds data instances that refer to the same real-world entity. In this paper we examine applying deep learning (DL) to EM, to understand DL's benefits and limitations. We review many DL solutions that have been developed for related matching tasks in text processing (e.g., entity linking, textual entailment, etc.). We categorize these solutions and define a space of DL solutions for EM, as embodied by four solutions with varying representational power: SIF, RNN, Attention, and Hybrid. Next, we investigate the types of EM problems for which DL can be helpful. We consider three such problem types, which match structured data instances, textual instances, and dirty instances, respectively. We empirically compare the above four DL solutions with Magellan, a state-of-the-art learning-based EM solution. The results show that DL does not outperform current solutions on structured EM, but it can significantly outperform them on textual and dirty EM. For practitioners, this suggests that they should seriously consider using DL for textual and dirty EM problems. Finally, we analyze DL's performance and discuss future research directions.
Both convolutional and recurrent operations are building blocks that process one local neighborhood at a time. In this paper, we present non-local operations as a generic family of building blocks for capturing long-range dependencies. Inspired by the classical non-local means method in computer vision, our non-local operation computes the response at a position as a weighted sum of the features at all positions. This building block can be plugged into many computer vision architectures. On the task of video classification, even without any bells and whistles, our non-local models can compete or outperform current competition winners on both Kinetics and Charades datasets. In static image recognition, our non-local models improve object detection/segmentation and pose estimation on the COCO suite of tasks. Code will be made available.
We present graph attention networks (GATs), novel neural network architectures that operate on graph-structured data, leveraging masked self-attentional layers to address the shortcomings of prior methods based on graph convolutions or their approximations. By stacking layers in which nodes are able to attend over their neighborhoods' features, we enable (implicitly) specifying different weights to different nodes in a neighborhood, without requiring any kind of costly matrix operation (such as inversion) or depending on knowing the graph structure upfront. In this way, we address several key challenges of spectral-based graph neural networks simultaneously, and make our model readily applicable to inductive as well as transductive problems. Our GAT models have achieved state-of-the-art results across three established transductive and inductive graph benchmarks: the Cora and Citeseer citation network datasets, as well as a protein-protein interaction dataset (wherein test graphs are entirely unseen during training).
The dominant sequence transduction models are based on complex recurrent or convolutional neural networks in an encoder-decoder configuration. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train. Our model achieves 28.4 BLEU on the WMT 2014 English-to-German translation task, improving over the existing best results, including ensembles by over 2 BLEU. On the WMT 2014 English-to-French translation task, our model establishes a new single-model state-of-the-art BLEU score of 41.0 after training for 3.5 days on eight GPUs, a small fraction of the training costs of the best models from the literature. We show that the Transformer generalizes well to other tasks by applying it successfully to English constituency parsing both with large and limited training data.
Many combinatorial optimization problems over graphs are NP-hard, and require significant specialized knowledge and trial-and-error to design good heuristics or approximation algorithms. Can we automate this challenging and tedious process, and learn the algorithms instead? In many real world applications, it is typically the case that the same type of optimization problem is solved again and again on a regular basis, maintaining the same problem structure but differing in the data. This provides an opportunity for learning heuristic algorithms which can exploit the structure of such recurring problems. In this paper, we propose a unique combination of reinforcement learning and graph embedding to address this challenge. The learned greedy policy behaves like a meta-algorithm which incrementally constructs a solution, and the action is determined by the output of a graph embedding network capturing the current state of the solution. We show that our framework can be applied to a diverse range of optimization problems over graphs, and provide evidence that our learning approach can compete with or outperform specialized heuristics or approximation algorithms for the Minimum Vertex Cover, Maximum Cut and Traveling Salesman Problems.
Supervised learning on molecules has incredible potential to be useful in chemistry, drug discovery, and materials science. Luckily, several promising and closely related neural network models invariant to molecular symmetries have already been described in the literature. These models learn a message passing algorithm and aggregation function to compute a function of their entire input graph. At this point, the next step is to find a particularly effective variant of this general approach and apply it to chemical prediction benchmarks until we either solve them or reach the limits of the approach. In this paper, we reformulate existing models into a single common framework we call Message Passing Neural Networks (MPNNs) and explore additional novel variations within this framework. Using MPNNs we demonstrate state of the art results on an important molecular property prediction benchmark, results we believe are strong enough to justify retiring this benchmark.
Knowledge bases play a crucial role in many applications, for example question answering and information retrieval. Despite the great effort invested in creating and maintaining them, even the largest representatives (e.g., Yago, DBPedia or Wikidata) are highly incomplete. We introduce relational graph convolutional networks (R-GCNs) and apply them to two standard knowledge base completion tasks: link prediction (recovery of missing facts, i.e. subject-predicate-object triples) and entity classification (recovery of missing attributes of entities). R-GCNs are a generalization of graph convolutional networks, a recent class of neural networks operating on graphs, and are developed specifically to deal with highly multi-relational data, characteristic of realistic knowledge bases. Our methods achieve competitive results on standard benchmarks for both tasks.
Evaluating similarity between graphs is of major importance in several computer vision and pattern recognition problems, where graph representations are often used to model objects or interactions between elements. The choice of a distance or similarity metric is, however, not trivial and can be highly dependent on the application at hand. In this work, we propose a novel metric learning method to evaluate distance between graphs that leverages the power of convolutional neural networks, while exploiting concepts from spectral graph theory to allow these operations on irregular graphs. We demonstrate the potential of our method in the field of connectomics, where neuronal pathways or functional connections between brain regions are commonly modelled as graphs. In this problem, the definition of an appropriate graph similarity function is critical to unveil patterns of disruptions associated with certain brain disorders. Experimental results on the ABIDE dataset show that our method can learn a graph similarity metric tailored for a clinical application, improving the performance of a simple k-nn classifier by 11.9% compared to a traditional distance metric.
Analyzing interconnection structures among underlying entities or objects in a dataset through the use of graph analytics has been shown to provide tremendous value in many application domains. However, graphs are not the primary representation choice for storing most data today, and in order to have access to these analyses, users are forced to extract data from their data stores, construct the requisite graphs, and then load them into some graph engine in order to execute their graph analysis task. Moreover, these graphs can be significantly larger than the initial input stored in the database, making it infeasible to construct or analyze such graphs in memory. In this paper we address both of these challenges by building a system that enables users to declaratively specify graph extraction tasks over a relational database schema and then execute graph algorithms on the extracted graphs. We propose a declarative domain-specific language for this purpose, and pair it up with a novel condensed, in-memory representation that significantly reduces the memory footprint of these graphs, permitting analysis of larger-than-memory graphs. We present a general algorithm for creating this condensed representation for a large class of graph extraction queries against arbitrary schemas. We observe that the condensed representation suffers from a duplication issue, that results in inaccuracies for most graph algorithms. We then present a suite of in-memory representations that handle this duplication in different ways and allow trading off the memory required and the computational cost for executing different graph algorithms. We introduce novel deduplication algorithms for removing this duplication in the graph, which are of independent interest for graph compression, and provide a comprehensive experimental evaluation over several real-world and synthetic datasets illustrating these trade-offs.
We present a scalable approach for semi-supervised learning on graph-structured data that is based on an efficient variant of convolutional neural networks which operate directly on graphs. We motivate the choice of our convolutional architecture via a localized first-order approximation of spectral graph convolutions. Our model scales linearly in the number of graph edges and learns hidden layer representations that encode both local graph structure and features of nodes. In a number of experiments on citation networks and on a knowledge graph dataset we demonstrate that our approach outperforms related methods by a significant margin.
A formal machine reader is a tool able to transform natural language text into formal structured knowledge so as the latter can be interpreted by machines, according to a shared semantics. FRED is a formal machine reader for the semantic web: its output is a RDF/OWL graph, whose design is based on frame semantics. FRED's graph are domain- and task-independent, making the tool suitable to be used as a semantic middleware for domain- or task-specific applications. To serve this purpose, it is available both as REST service and as Python library. This paper provides details about FRED's capabilities, design issues, implementation and evaluation.
Despite significant recent advances in the field of face recognition,
implementing face verification and recognition efficiently at scale presents
serious challenges to current approaches. In this paper we present a system,
called FaceNet, that directly learns a mapping from face images to a compact
Euclidean space where distances directly correspond to a measure of face
similarity. Once this space has been produced, tasks such as face recognition,
verification and clustering can be easily implemented using standard techniques
with FaceNet embeddings as feature vectors.
Our method uses a deep convolutional network trained to directly optimize the
embedding itself, rather than an intermediate bottleneck layer as in previous
deep learning approaches. To train, we use triplets of roughly aligned matching
/ non-matching face patches generated using a novel online triplet mining
method. The benefit of our approach is much greater representational
efficiency: we achieve state-of-the-art face recognition performance using only
128-bytes per face.
On the widely used Labeled Faces in the Wild (LFW) dataset, our system
achieves a new record accuracy of 99.63%. On YouTube Faces DB it achieves
95.12%. Our system cuts the error rate in comparison to the best published
result by 30% on both datasets.
This tutorial brings together perspectives on ER from a variety of fields, including databases, machine learning, natural language processing and information retrieval, to provide, in one setting, a survey of a large body of work. We discuss both the practical aspects and theoretical underpinnings of ER. We describe existing solutions, current challenges, and open research problems.
In this article, we address the problem of defining scalable kernels on large graphs with discrete node labels. Key to our approach is the Weisfeiler-Lehman test of isomorphism, which allows us to compute a sequence of graphs which capture the topological and label information of the original graph in a runtime which is linear in the number of edges. We can apply existing graph kernels on this graph sequence and make them take into account the structural information which they ignored before. We can also define new, efficient graph kernels: In particular, a subtree kernel whose runtime is linear in the number of edges in the input graphs and in the maximum height of the subtrees considered.
In this paper we demonstrate that finite linear combinations of compositions of a fixed, univariate function and a set of
affine functionals can uniformly approximate any continuous function ofn real variables with support in the unit hypercube; only mild conditions are imposed on the univariate function. Our results
settle an open question about representability in the class of single hidden layer neural networks. In particular, we show
that arbitrary decision regions can be arbitrarily well approximated by continuous feedforward neural networks with only a
single internal, hidden layer and any continuous sigmoidal nonlinearity. The paper discusses approximation properties of other
possible types of nonlinearities that might be implemented by artificial neural networks.
Many underlying relationships among data in several areas of science and engineering, e.g., computer vision, molecular chemistry, molecular biology, pattern recognition, and data mining, can be represented in terms of graphs. In this paper, we propose a new neural network model, called graph neural network (GNN) model, that extends existing neural network methods for processing the data represented in graph domains. This GNN model, which can directly process most of the practically useful types of graphs, e.g., acyclic, cyclic, directed, and undirected, implements a function tau(G,n) isin IR
m
that maps a graph G and one of its nodes
n
into an
m
-dimensional Euclidean space. A supervised learning algorithm is derived to estimate the parameters of the proposed GNN model. The computational cost of the proposed algorithm is also considered. Some experimental results are shown to validate the proposed learning algorithm, and to demonstrate its generalization capabilities.
A Direct Mapping of Relational Data to RDF
- M Arenas
- A Bertails
- E Prud'hommeaux
- J Sequeda
M. Arenas, A. Bertails, E. Prud'hommeaux, and J. Sequeda. A Direct Mapping of Relational Data to RDF.
https://www.w3.org/TR/rdb-direct-mapping/, September 2012.
Matching entities across different knowledge graphs with graph embeddings
- M Azmy
- P Shi
- J Lin
- I F Ilyas
M. Azmy, P. Shi, J. Lin, and I. F. Ilyas. Matching entities across different knowledge graphs with graph embeddings. CoRR, abs/1903.06607, 2019.
Building dynamic knowledge graphs from text using machine reading comprehension
- R Das
- T Munkhdalai
- X Yuan
- A Trischler
- A Mccallum
R. Das, T. Munkhdalai, X. Yuan, A. Trischler, and A. McCallum. Building dynamic knowledge graphs from text
using machine reading comprehension. arXiv preprint arXiv:1810.05682, 2018.
BERT: pre-training of deep bidirectional transformers for language understanding
- J Devlin
- M Chang
- K Lee
- K Toutanova
J. Devlin, M. Chang, K. Lee, and K. Toutanova. BERT: pre-training of deep bidirectional transformers for
language understanding. In Proceedings of NAACL-HLT, pages 4171-4186, 2019.
Fast record linkage for company entities
- T Gschwind
- C Miksovic
- J Minder
- K Mirylenka
- P Scotton
T. Gschwind, C. Miksovic, J. Minder, K. Mirylenka, and P. Scotton. Fast record linkage for company entities.
IEEE Big Data 2019, December 9-12, 2019, Los Angeles, CA, USA, 2019.
Attention, learn to solve routing problems
- W Kool
- H Van Hoof
- M Welling
W. Kool, H. van Hoof, and M. Welling. Attention, learn to solve routing problems! In International Conference
on Learning Representations, 2019.
Resnet with one-neuron hidden layers is a universal approximator
- H Lin
- S Jegelka
H. Lin and S. Jegelka. Resnet with one-neuron hidden layers is a universal approximator. In NeurIPS 2018,
2018.
Similarity learning with higher-order proximity for brain network analysis
- G Ma
- N K Ahmed
- T L Willke
- D Sengupta
- M W Cole
- N B Turk-Browne
- P S Yu
G. Ma, N. K. Ahmed, T. L. Willke, D. Sengupta, M. W. Cole, N. B. Turk-Browne, and P. S. Yu. Similarity
learning with higher-order proximity for brain network analysis. CoRR, abs/1811.02662, 2018.
Applicability of latent dirichlet allocation for company modeling
- K Mirylenka
- C Miksovic
- P Scotton
K. Mirylenka, C. Miksovic, and P. Scotton. Applicability of latent dirichlet allocation for company modeling. In
Industrial Conference on Data Mining (ICDM'2016), 2016.
Recurrent neural networks for modeling company-product time series
- K Mirylenka
- C Miksovic
- P Scotton
K. Mirylenka, C. Miksovic, and P. Scotton. Recurrent neural networks for modeling company-product time
series. Proceedings of the workshop on Advanced Analytics and Learning on Temporal Data (AALTD) in conjunction with ECML PKDD, pages 29-36, 2016.