Palash Goyal

Graph embedding techniques, applications, and performance: A survey

Abstract Graphs, such as social networks, word co-occurrence networks, and communication networks, occur naturally in various real-world applications. Analyzing them yields insight into the structure of society, language, and different patterns of communication. Many approaches have been proposed to perform the analysis. Recently, methods which use the representation of graph nodes in vector space have gained traction from the research community. In this survey, we provide a comprehensive and structured analysis of various graph embedding techniques proposed in the literature. We first introduce the embedding task and its challenges such as scalability, choice of dimensionality, and features to be preserved, and their possible solutions. We then present three categories of approaches based on factorization methods, random walks, and deep learning, with examples of representative algorithms in each category and analysis of their performance on various tasks. We evaluate these state-of-the-art methods on a few common datasets and compare their performance against one another. Our analysis concludes by suggesting some potential applications and future directions. We finally present the open-source Python library we developed, named GEM (Graph Embedding Methods, available at https://github.com/palash1992/GEM ), which provides all presented algorithms within a unified interface to foster and facilitate research on the topic.

Future Automation Engineering using Structural Graph Convolutional Neural Networks.

The digitalization of automation engineering generates large quantities of engineering data that is interlinked in knowledge graphs. Classifying and clustering subgraphs according to their functionality is useful to discover functionally equivalent engineering artifacts that exhibit different graph structures. This paper presents a new graph learning algorithm designed to classify engineering data artifacts -- represented in the form of graphs -- according to their structure and neighborhood features. Our Structural Graph Convolutional Neural Network (SGCNN) is capable of learning graphs and subgraphs with a novel graph invariant convolution kernel and downsampling/pooling algorithm. On a realistic engineering-related dataset, we show that SGCNN is capable of achieving ~91% classification accuracy.

GEM: A Python package for graph embedding methods

Many physical systems in the world involve interactions between different entities and can be represented as graphs. Understanding the structure and analyzing properties of graphs are hence paramount to developing insights into the physical systems. Graph embedding, which aims to represent a graph in a low dimensional vector space, takes a step in this direction. The embeddings can be used for various tasks on graphs such as visualization, clustering, classification and prediction.

Rapid Data Integration and Analysis for Upstream Oil and Gas Applications

The increasingly large number of sensors and instruments in the oil and gas industry, along with novel means of communication in the enterprise has led to a corresponding increase in the volume of data that is recorded in various information repositories. The variety of information sources is also expanding: from traditional relational databases to time series data, social network communications, collections of unsorted text reports, and linked data available on the Web. Enabling end-to-end optimization considering these diverse types of information requires creating semantic links between them. Though integration of data across silo-ed databases has been recognized as a problem for a long time, it has proven to be difficult to accomplish due to the complexity of the data arrangement within databases, scarcity of metadata that describe the content, lack of a direct mapping between related entities across databases, and the several types of data represented within a database. In addition, there are large amounts of unstructured text data such as text entries in databases and document repositories. These contain valuable information on processes from the field but there is currently no method to convert this raw data to useable information. The Center for Interactive Smart Oilfield Technologies (CiSoft) is a USC-Chevron Center of Excellence for Research and Academic Training on Smart Oilfield Technologies. We describe the Integrated Optimization project at CiSoft which has the goal of developing a framework for automated linking of heterogeneous data sources and analysis of the integrated data in the context of upstream applications.