6,091 Hits in 5.2 sec

Multi-Label Patent Categorization with Non-Local Attention-Based Graph Convolutional Network

Pingjie Tang, Meng Jiang, Bryan (Ning) Xia, Jed W. Pitera, Jeffrey Welser, Nitesh V. Chawla
In this work, we propose a label attention model based on graph convolutional network.  ...  On a large CIRCA patent database, we evaluate the performance of our model and as many as seven competitive baselines.  ...  AnnexML (Tagami 2017 ) presents a graph embedding method to cope with several limitations of SLEEC.  ... 
doi:10.1609/aaai.v34i05.6435 fatcat:z25s27ojf5f7vnbh7bgmjpdziu

Subset Node Representation Learning over Large Dynamic Graphs [article]

Xingzhi Guo, Baojian Zhou, Steven Skiena
2021 arXiv   pre-print
Dynamic graph representation learning is a task to learn node embeddings over dynamic networks, and has many important applications, including knowledge graphs, citation networks to social networks.  ...  Based on recent advances in local node embedding and a novel computation of dynamic personalized PageRank vector (PPV), DynamicPPE has two key ingredients: 1) the per-PPV complexity is 𝒪(m d̅ / ϵ) where  ...  Patent (US Patent Graph) The citation network of US patent[14] contains 2,738,011 nodes with 13,960,811 citations range from year 1963 to 1999.  ... 
arXiv:2106.01570v1 fatcat:iegarl2fsbhutb4sesspfab6he

Guest Editorial: Tech Mining for Engineering Management: An Introduction

Yi Zhang, Ying Huang, Denise Chiavetta, Alan L. Porter
2021 IEEE transactions on engineering management  
Competitors' Innovation Activities: Analyzing the Competitive Patent Landscape Based on Semantic Anchor Points, in which the authors specifically focused on the activities of patent assignees and proposed  ...  CTI-Oriented Patent Analysis Patent data have long been one of the key ST&I sources for gaining CTI, and facilitating patent statistics, including unique attributes [e.g., international patent classification  ... 
doi:10.1109/tem.2021.3061862 fatcat:kudqaewka5hd5p5qvq3sqqvfgy

Simple and Effective Graph Autoencoders with One-Hop Linear Models [article]

Guillaume Salha, Romain Hennequin, Michalis Vazirgiannis
2020 arXiv   pre-print
Over the last few years, graph autoencoders (AE) and variational autoencoders (VAE) emerged as powerful node embedding methods, with promising performances on challenging tasks such as link prediction  ...  Graph AE, VAE and most of their extensions rely on multi-layer graph convolutional networks (GCN) encoders to learn vector space representations of nodes.  ...  Existing models usually rely on graph neural networks (GNN) to encode nodes into embeddings.  ... 
arXiv:2001.07614v3 fatcat:bpe3e5ms7nbdxajpskmqxmpwza

A Comparison between Political Claims Analysis and Discourse Network Analysis: The Case of Software Patents in the European Union

Philip Leifeld, Sebastian Haunss
2010 Social Science Research Network  
We compare discourse network analysis with political claims analysis, a competing method, and apply both methods to the European-level discourse on software patents.  ...  It is based on social network analysis and qualitative content analysis and takes an entirely relational perspective.  ...  The approach called discourse network analysis is compared with a competing method, political claims analysis, and both methods are applied to the European-level discourse on software patents.  ... 
doi:10.2139/ssrn.1617194 fatcat:xqc2g6vvljbxfoll52fa7qlv6m

A Degeneracy Framework for Scalable Graph Autoencoders [article]

Guillaume Salha, Romain Hennequin, Viet Anh Tran, Michalis Vazirgiannis
2019 arXiv   pre-print
We evaluate and discuss our method on several variants of existing graph AE and VAE, providing the first application of these models to large graphs with up to millions of nodes and edges.  ...  We achieve empirically competitive results w.r.t. several popular scalable node embedding methods, which emphasizes the relevance of pursuing further research towards more scalable graph AE and VAE.  ...  Graph AE/VAE training All graph AE/VAE models were trained on 200 epochs to return 16-dim embeddings, except for Patent (500 epochs, 32dim).  ... 
arXiv:1902.08813v2 fatcat:xnw6my5ekrcadj7xay5az5qoaq

Trans4E: Link Prediction on Scholarly Knowledge Graphs

Mojtaba Nayyeri, Gokce Muge Cil, Sahar Vahdati, Francesco Osborne, Mahfuzur Rahman, Simone Angioni, Angelo Salatino, Diego Reforgiato Recupero, Nadezhda Vassilyeva, Enrico Motta, Jens Lehmann
2021 Neurocomputing  
In recent years, link prediction approaches based on Knowledge Graph Embedding models became the first aid for this issue.  ...  Trans4E was applied on two large-scale knowledge graphs, the Academia/Industry DynAmics (AIDA) and Microsoft Academic Graph (MAG), for completing the information about Fields of Study (e.g., 'neural networks  ...  Acknowledgements We gratefully acknowledge the support of NVIDIA Corporation with the donation of the Titan X GPU used for this research.  ... 
doi:10.1016/j.neucom.2021.02.100 fatcat:j4iuey3rzreppow2zt3qnusjvm

AIDA: a Knowledge Graph about Research Dynamics in Academia and Industry

Simone Angioni, Angelo Salatino, Francesco Osborne, Diego Reforgiato Recupero, Enrico Motta
2021 Quantitative Science Studies  
We evaluated the different parts of the generation pipeline on a manually crafted gold standard yielding competitive results.  ...  In this paper, we introduce the Academia/Industry DynAmics (AIDA) Knowledge Graph, which describes 21M publications and 8M patents according to the research topics drawn from the Computer Science Ontology  ...  We also thank Dimensions for sharing their large dataset of patents.  ... 
doi:10.1162/qss_a_00162 fatcat:diw25svrw5hlvddin4truogq3y

Information Retrieval and Visualization for Searching Scientific articles and Patents

Lipika Dey, Hemant Gupta, Kunal Ranjan
2015 Research in Computing Science  
However, search is a humandriven activity and the result of such analysis is largely dependent on the initial inputs that are provided by the expert.  ...  We present results of experiments based on research abstracts made available by digital libraries and US patent office.  ...  Figure 5 presents a graph, each of whose nodes are topics that represent the area of Wireless Sensor Networks, which is the same as the topic-evolution graph component for the area, with one difference  ... 
doi:10.13053/rcs-90-1-21 fatcat:hsqovp6wrbaaziioy5a7xm4zxe

The Impact of Co-Inventor Networks on Smart Cleantech Innovation: The Case of Montreal Agglomeration

Ekaterina Turkina, Boris Oreshkin
2021 Sustainability  
We use patent big data and apply a combination of network analysis techniques to explore the social structure of the Montreal tech community and its embeddedness in the global innovation landscape.  ...  Our analysis reveals the importance of both local and international ties for the general development of innovations in Montreal's competitive urban economy, with a stronger impact of international ties  ...  Citation Networks and Co-Inventor Networks The network analysis is based on the graphs defined by edges and vertices. Throughout the analysis we define a vertex to be a geolocated patent inventor.  ... 
doi:10.3390/su13137270 fatcat:gonmf354nzennp72doj5z7cksy

Network Analysis of Innovation in the Internet of Things

Fumihiko Isada, Yuriko Isada
2018 Interdisciplinary Description of Complex Systems  
Conclusion: Drawing on an intellectual property database and employing social network analysis, this research quantifies the structure of innovation networks in terms of the results and operational efficiency  ...  Method: In this research, the relationship between the network structure and the result of innovation was analysed through social network analysis.  ...  In graph theory and network analysis, indicators of centrality identify the most important vertices within a graph.  ... 
doi:10.7906/indecs.16.2.2 fatcat:3taflcmwwzfnhnjl6eioihrnva

IP Analytics and Machine Learning Applied to Create Process Visualization Graphs for Chemical Utility Patents

Amy J. C. Trappey, Charles V. Trappey, Chih-Ping Liang, Hsin-Jung Lin
2021 Processes  
A computer-supported graph-based knowledge representation interface is developed to plot the extracted chemical terms and their chemical process links as a network of nodes with connecting arcs.  ...  The researcher must be able to understand the latest patents and literature in order to develop new chemicals and processes that do not infringe on existing claims and processes.  ...  For easy analysis of complex network data and simulation models, NetworkX has algorithms for complex network analysis and graph visualization.  ... 
doi:10.3390/pr9081342 fatcat:fbh4kxol5vgidnv2qs7qa5pcxa


Matej Babič
2018 Journal of Production Engineering  
In this article we present new method for pattern recognition and calculating all closeness centralization of network created with connection maximum value of 3D graph of hardenend specimens.  ...  We use a method which combines an intelligent genetic algorithm and multiple regression to predict all closeness centralization of network created with connection maximum value of 3D graph of hardened  ...  Since its inception twenty years ago, GP has been used to solve a wide range of practical problems, producing a number of human-competitive results and even patentable new inventions.  ... 
doi:10.24867/jpe-2018-01-021 fatcat:ee5jl2vtnvejpdz7l6wqgah3tm

Learning to Identify High Betweenness Centrality Nodes from Scratch: A Novel Graph Neural Network Approach [article]

Changjun Fan, Li Zeng, Yuhui Ding, Muhao Chen, Yizhou Sun, Zhong Liu
2019 arXiv   pre-print
In this paper, we focus on identifying nodes with high BC in a graph, since many application scenarios are built upon retrieving nodes with top-k BC.  ...  Betweenness centrality (BC) is one of the most used centrality measures for network analysis, which seeks to describe the importance of nodes in a network in terms of the fraction of shortest paths that  ...  Each edge connects an author to one of his publications. cit-Patents is a citation network of U.S. patents. Nodes are patents and edges represent citations.  ... 
arXiv:1905.10418v4 fatcat:m2p7mcwdmjfjbip2fjvzvzf5qe

Social Network Analysis of the Irish Biotech Industry: Implications for Digital Ecosystems [chapter]

Chris van Egeraat, Declan Curran
2010 Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering  
The study applies social network analysis to determine the structure of networks of company directors and inventors in the biotech sector.  ...  This paper presents an analysis of the socio-spatial structures of innovation, collaboration and knowledge flow among SMEs in the Irish biotech sector.  ...  One established framework for analysing network structure is that of "small world" network analysis. Small world analysis is concerned with the density and reach of ties.  ... 
doi:10.1007/978-3-642-14859-0_3 fatcat:77vxtabxtjafpnhcefbgnsna44
« Previous Showing results 1 — 15 out of 6,091 results