269 Hits in 10.9 sec

Every Child Should Have Parents: A Taxonomy Refinement Algorithm Based on Hyperbolic Term Embeddings

Rami Aly, Shantanu Acharya, Alexander Ossa, Arne Köhn, Chris Biemann, Alexander Panchenko
2019 Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics  
(pre-induced) taxonomy as well as for attaching disconnected terms in a taxonomy.  ...  We introduce the use of Poincaré embeddings to improve existing state-of-the-art approaches to domain-specific taxonomy induction from text as a signal for both relocating wrong hyponym terms within a  ...  We also thank three anonymous reviewers and Simone Paolo Ponzetto for providing useful feedback on this work.  ... 
doi:10.18653/v1/p19-1474 dblp:conf/acl/AlyAOKBP19 fatcat:26rjvtuv7vbtzgomq337hw5ary

HyperExpan: Taxonomy Expansion with Hyperbolic Representation Learning [article]

Mingyu Derek Ma, Muhao Chen, Te-Lin Wu, Nanyun Peng
2021 arXiv   pre-print
In this paper, we present HyperExpan, a taxonomy expansion algorithm that seeks to preserve the structure of a taxonomy in a more expressive hyperbolic embedding space and learn to represent concepts and  ...  with the geometric properties of a hyperbolic space.  ...  Hyperbolic Representation Learning Nickel and Kiela (2017) present an efficient algorithm to learn embeddings in a supervised manner based on Riemannian optimization and shows it performs well on link  ... 
arXiv:2109.10500v1 fatcat:7uupuxk3ozd4bkjrqndnai7zoq

Knowledge Representation and Rule Mining in Entity-Centric Knowledge Bases [chapter]

Fabian M. Suchanek, Jonathan Lajus, Armand Boschin, Gerhard Weikum
2019 Lecture Notes in Computer Science  
In this paper, we first discuss the knowledge representation that has emerged as a pragmatic consensus in the research community of entity-centric knowledge bases.  ...  Then, we describe how these knowledge bases can be mined for logical rules. Finally, we discuss how entities can be represented alternatively as vectors in a vector space, by help of neural networks.  ...  Or we could notice that, if, in addition, A is the parent of some child, then B is usually also a parent of that child (although not always).  ... 
doi:10.1007/978-3-030-31423-1_4 fatcat:ggtrgoced5dwxis6mlrtmzivui

Ontology visualization methods---a survey

Akrivi Katifori, Constantin Halatsis, George Lepouras, Costas Vassilakis, Eugenia Giannopoulou
2007 ACM Computing Surveys  
Ontologies, as sets of concepts and their interrelations in a specific domain, have proven to be a useful tool in the areas of digital libraries, the semantic web, and personalized information management  ...  As a result, there is a growing need for effective ontology visualization for design, management and browsing.  ...  The hyperbolic tree technique is based on a hyperbolic transformation.  ... 
doi:10.1145/1287620.1287621 fatcat:pck5kq55pvfsvf4nw7rkkotdpu

Graphics processing unit (GPU) programming strategies and trends in GPU computing

André R. Brodtkorb, Trond R. Hagen, Martin L. Sætra
2013 Journal of Parallel and Distributed Computing  
Through the scientific papers in this thesis, we present efficient hardware-adapted shallow water simulations on the GPU, based on a high-resolution centralupwind scheme.  ...  The topics range from best practices for stencil computations on the GPU to adaptive mesh refinement.  ...  Because each child grid has twice the resolution of its parent, one should in principle use two time steps on the child grid for each time step on the parent grid.  ... 
doi:10.1016/j.jpdc.2012.04.003 fatcat:7s4fnkx3yrekbmxabztmto5fzq

Learning Representations For Images With Hierarchical Labels [article]

Ankit Dhall
2020 arXiv   pre-print
Although, contrasting in nature, both the CNN-classifiers injected with hierarchical information, and the embedding-based models outperform a hierarchy-agnostic model on the newly presented, real-world  ...  Taking a step further in this direction, we model more explicitly the label-label and label-image interactions by using order-preserving embedding-based models, prevalent in natural language, and tailor  ...  F (I) = p(·) where, p(·) is the conditional probability for every child node given the parent, ). F takes as input image I.  ... 
arXiv:2004.00909v2 fatcat:hxreipp6qzgszbbl6zpmlcsczy

Hyperbolic Disentangled Representation for Fine-Grained Aspect Extraction [article]

Chang-You Tai, Ming-Yao Li, Lun-Wei Ku
2021 arXiv   pre-print
Second, each seed words representation should have different latent semantics and be distinct when it represents a different aspect.  ...  In this paper, we propose HDAE, a hyperbolic disentangled aspect extractor in which a hyperbolic aspect classifier captures words latent hierarchies, and aspect-disentangled representation models the distinct  ...  Every child should have parents: A (AAAI). taxonomy refinement algorithm based on hyperbolic term Hu, L.; Xu, S.; Li, C.; Yang, C.; Shi, C.; Duan, N.; Xie, X.; embeddings. arXiv preprint  ... 
arXiv:2112.09215v1 fatcat:rzrh35r4frentiw4dxlu66zzoa

Learning What You Need from What You Did: Product Taxonomy Expansion with User Behaviors Supervision [article]

Sijie Cheng, Zhouhong Gu, Bang Liu, Rui Xie, Wei Wu, Yanghua Xiao
2022 arXiv   pre-print
Taxonomies have been widely used in various domains to underpin numerous applications.  ...  However, taxonomies need to constantly capture the newly emerged terms or concepts in e-commerce platforms to keep up-to-date, which is expensive and labor-intensive if it relies on manual maintenance  ...  These two types of position embedding, p parent and p child , indicate the node identity of the parent or child.  ... 
arXiv:2203.14921v1 fatcat:lxy5sp3zajft5hjlj6dgbv55lu

Shape recovery algorithms using level sets in 2-D/3-D medical imagery: a state-of-the-art review

J.S. Suri, Kecheng Liu, S. Singh, S.N. Laxminarayan, Xiaolan Zeng, L. Reden
2002 IEEE Transactions on Information Technology in Biomedicine  
In this class, we present four kinds: clustering-based, Bayesian bidirectional classifier-based, shape-based and coupled constrained-based.  ...  The paper then focuses on the first core class of level sets, known as "level sets without regularizers."  ...  A binary heap had the property that the value at a given "child" position was always larger than or equal to the value at its "parent" position .  ... 
doi:10.1109/4233.992158 pmid:11936600 fatcat:7k2degbqmzcjhp7ssykjtkjo5e

EncCon: An Approach to Constructing Interactive Visualization of Large Hierarchical Data

Quang Vinh Nguyen, Mao Lin Huang
2005 Information Visualization  
Our technique uses a rectangular division algorithm for recursively positioning the graph.  ...  Besides the layout algorithm, EncCon uses a new focus þ context viewing technique for the navigation of large hierarchies.  ...  If (m,n) is a directed edge in T, we then say m is the parent of n and n is a child of m. A leaf is a vertex with no children.  ... 
doi:10.1057/palgrave.ivs.9500087 fatcat:gedbvujv55ax7lpq53zj3xkrn4

From Keyword Search to Exploration: Designing Future Search Interfaces for the Web

Max L. Wilson
2010 Foundations and Trends® in Web Science  
with and have control over their findings.  ...  Search is one of the keys to the Web's success. The elegant way in which search results are returned has been well researched and is usually remarkably effective. However,  ...  They found that recall improved when searchers were allowed to augment their queries with terms from a thesaurus generated via a clustering-based algorithm.  ... 
doi:10.1561/1800000003 fatcat:jpiox47bxneefmgrajbhosfvuq

Multilayer Dense Connections for Hierarchical Concept Classification [article]

Toufiq Parag, Hongcheng Wang
2021 arXiv   pre-print
In particular, when a CNN based image classifier correctly identifies the image of a Chimpanzee, its output does not clarify that Chimpanzee is a member of Primate, Mammal, Chordate families and a living  ...  Multinomial logistic regression with a single final layer of dense connections has become the ubiquitous technique for CNN-based classification.  ...  It is important to distinguish our work from the hyperbolic embedding studies [32, 27] . Khrulkov et al.  ... 
arXiv:2003.09015v2 fatcat:lc6pnu56o5fy7mydwczfxgkjeu

A Survey of Radial Methods for Information Visualization

G.M. Draper, Y. Livnat, R.F. Riesenfeld
2009 IEEE Transactions on Visualization and Computer Graphics  
A taxonomy for radial visualization is proposed, in the form of seven design patterns encompassing nearly all recent work in this area.  ...  It is hoped that our taxonomy will provide a framework for facilitating discourse among researchers and stimulate the development of additional theories and systems involving radial visualization as a  ...  The user can refine a query, bubbles which are rendered tangentially to a key word on the parent bubble's circumference.  ... 
doi:10.1109/tvcg.2009.23 pmid:19590103 fatcat:ch3edindivcw7pxhz6xommrrdi

Machine Knowledge: Creation and Curation of Comprehensive Knowledge Bases [article]

Gerhard Weikum, Luna Dong, Simon Razniewski, Fabian Suchanek
2021 arXiv   pre-print
Over the last decade, large-scale knowledge bases, also known as knowledge graphs, have been automatically constructed from web contents and text sources, and have become a key asset for search engines  ...  It covers models and methods for discovering and canonicalizing entities and their semantic types and organizing them into clean taxonomies.  ...  It is a great pleasure and honor to have such wonderful colleagues in our research community.  ... 
arXiv:2009.11564v2 fatcat:vh2lqfmhhbcwpf6dcsej3hhvgy

Towards the Improvement of Automated Scientific Document Categorization by Deep Learning [article]

Thomas Krause
2017 arXiv   pre-print
A design will be proposed using a convolutional neural network as a classifier and integrating this into a REST based API.  ...  using REST based services.  ...  , see also section 1.1), where membership in a child category also implies membership in the parent category (and so on).  ... 
arXiv:1706.05719v1 fatcat:chnkh3rrzvevrlejjlzlgwsg3u
« Previous Showing results 1 — 15 out of 269 results